Sussex Migration Working Papers Harnessing the Potential of Migration and Return to Promote Development: Applying Concepts to West Africa Savina Ammassari Richard Black July 2001

2

Table of contents

1. Introduction ........................................................................................................................................4 2. Contemporary trends in migration: the case of West Africa ....................................................................6 3. Consequences of international migration..............................................................................................7

3.1 Labour market effects ....................................................................................................................8 3.2 Migrant remittances .......................................................................................................................9 3.3 The brain drain............................................................................................................................ 10

4. The study of return and transnational migration: methodological issues............................................... 11

4.1 Problems of definition and measurement....................................................................................... 12 4.2 Characteristics of migrants and multiple selectivity......................................................................... 13 4.3 Types of migration and return ...................................................................................................... 14 4.4 Return motives and readjustment problems .................................................................................. 16

5 Development implications of different forms of capital transfer.............................................................16

5.1 Financial capital ........................................................................................................................... 17 5.2 Human capital ............................................................................................................................. 18 5.3 Social capital ............................................................................................................................... 19

6. Elements for a meso-level approach to migration, return and development ........................................... 20

6.1 Survival and mobility strategies .................................................................................................... 21 The migrants’ experience abroad.................................................................................................... 22 The migrants’ experience on return. ............................................................................................... 23

6.2 Structural and institutional constraints .......................................................................................... 23 6.3 Migration, return and social networks ........................................................................................... 24

7. Linking micro and macro perspectives................................................................................................. 25

7.1 Methods...................................................................................................................................... 25 7.2 Policy .......................................................................................................................................... 26

Bibliography .......................................................................................................................................... 27

3

List of acronyms and abbreviations ENSSEA Ecôle Nationale Supérieure de Statistique et d’Economie Appliquée ILO International Labour Organization IOM International Organization for Migration ISSER Institute of Statistical, Social and Economic Research GNP Gross national product LSMS Living Standards Measurements Study NGO Non-governmental organisation ODA Official development assistance OECD Organization for Economic Co-operation and Development TOKTEN Transfer of Knowledge through Expatriate Nationals UN United Nations UNCTAD United Nations Conference on Trade and Development UNDP United Nations Development Programme

Preface This paper is the first in a series to be written as part of a research project on ‘Transnational Migration, Return and Development in West Africa’, funded by the Department for International Development and carried out by the Sussex Centre for Migration Research in collaboration with the Ecôle Nationale Supérieure de Statistique et d’Economie Appliquée (ENSSEA), Abidjan, the Institute of Statistical, Social and Economic Research (ISSER), University of Legon, Ghana.

4

1. Introduction It is generally recognised that international migration and development are closely interconnected. International migration is both a dependent and an independent variable of development (Fischer et al., 1997). Two questions have therefore kept the interest of migration scholars and policy makers: how is migration influenced by development and how does migration impact on development. To answer these questions particular attention has been focused on determinants of migration and its consequences in the areas of origin and destination as well as on the volume, patterns, and dynamics of migration. Yet the understanding of the relationship between migration and development remains limited because both are highly dynamic and complex processes, which vary across time and space and can be considered from very different perspectives (Appleyard, 1989; 1992; Papademetriou and Martin, 1991).

The need to develop a thorough understanding of the migration-development linkage is essential in an era characterised by increasing human mobility. It is generally assumed that today more than 130 million people live outside the country of their birth, a number estimated to be growing at a rate of 2% per year (World Bank, 2000). Human mobility is likely to assume even greater significance in the future. With increased transnational capital movements and the globalisation of trade in goods and services, people too can be expected to move across national boundaries with a greater frequency than ever before in what has been described as the ‘age of migration’ (Castles and Miller, 1998).

The total volume of South to North migration has grown particularly fast in the past forty years, linked to the widening gap in living standards between developing countries and economically more advanced countries (IOM, 1991; OECD, 2000). The most important increases have been from Southern Asia, Western Asia, and North Africa. But sub-Saharan Africa is also increasingly concerned by emigration to wealthier countries in the North due to its worsening economic, social, and political situation1. Western Europe and North

1 Overviews of South-North migration in the Asia-Pacific region and Africa can be found in Stahl (1991) and Adepoju (1991a) respectively.

America are the preferred destinations, along with oil-rich Gulf states2.

South to North migration has been the focus of much controversial debate since the 1960s. One central question has been whether it has a positive or negative impact on developing countries. Migration studies conducted in different parts of the world to assess the developmental impacts of migration have produced quite contradictory results, which make it difficult to draw any meaningful general conclusions. The problem is partly due to the fact that the theoretical base for understanding these forces remains weak (Massey et al., 1993). The contradictory nature of empirical findings also results from the fact that few studies distinguish the impacts of migration and return according to the many variables influencing these two complex phenomena.

It has been argued that emigration can alleviate some of the problems facing developing countries. Emigration may relieve labour market pressure and generate remittances, which constitute an important source of foreign exchange and income for migrants’ families. Financial and human capital transfers occurring through return migration can have a positive impact as they help improve the quality of life back home and promote socio-economic development. On the other hand, it has been stressed that emigration can also hamper development of the sending country because the most dynamic and ambitious people have a greater propensity to go abroad. The loss of skilled manpower – the so-called brain drain – may have detrimental effects as it weakens human capital. Also, migrant remittances do not necessarily have a positive developmental impact on the sending country as they may be used for consumption rather than investment purposes, or they may generate a problem of inflation.

2 The immigrant stock in Western Europe and North

America has been growing at 2.5%, a higher growth rate than the population growth rate (World Bank, 2000). Increases in the stock and proportion of migrants have been more significant in the North than in the South. The greatest increases have been registered in Western Europe (from 3% in 1965 to over 5% in 1990), whereas North America remains among the regions with the highest proportion of foreign-born residents (Tapinos and Delaunay, 2000).

5

This long-standing debate is yet to be terminated. In the case of the ‘brain drain’, for instance, while certain scholars and policy makers have emphasised its negative effects, others have argued that the problem is being overstated or have stressed the positive effects of the redistribution of human capital for the economy of both the receiving and sending country (Glaser, 1978). There also has been little consensus regarding the effects of emigration on labour markets in sending countries. Liberal economic theory considers it as a beneficial development enhancing mechanism that helps easing un- and underemployment. But it also has been argued that emigration can produce harmful labour shortages. Similarly, the discussion regarding the transfer and use of remittances remains unresolved. Although the importance of the volume of remittances is generally acknowledged, their unproductive use is often emphasised.

Return is another part of the migration process that has been studied with the aim to determine its consequences for the development of the emigration country. Attention focused on return migration so far has been limited, but here too there is little agreement regarding its impact. Emphasis has been set on the positive consequences of financial capital and human capital transferred back home by returning migrants. The first is brought back in the form of savings, whereas the second materialises through the training and work experience that migrants acquire in the host country. On the other hand, it has been argued that many migrants gain very little human capital, because they mainly do unskilled work, which does not teach them anything. It is moreover claimed that even if migrants were to acquire new skills abroad, it is unlikely that these can easily be used productively back home.

The existing body of theoretical and empirical evidence shows that the impact of international migration and return on development varies considerably, depending among other things on the volume, type, and timing of migration flows. The characteristics of migrants, degree and direction of selectivity, and situation of the countries involved in migration are also critical factors, which need to be taken into consideration. The consequences of migration also vary according to the level of analysis that is selected. There can be very different implications for individual migrants, their families, communities, and home countries. The short-term

effects of international migration may moreover differ significantly from its long-term effects.

Empirical evidence concerning migration and return gathered so far is too fragmentary and contradictory to allow us to draw clear conclusions and devise concrete policy measures. It is therefore crucial to examine the implications of migration for the meeting of international development targets in specific country settings, possibly allowing for regional comparisons, to gain a clearer and more comprehensive picture. While empirical evidence from southern Europe, Asia and the Pacific, the Middle East and the Caribbean is accumulating, less is known about the developmental impact of international migration and return in sub-Saharan Africa, despite the urgent need to alleviate poverty and foster sustainable development in this part of the world.

The purpose of this paper is to provide a state-of-the-art literature review and to stimulate discussions around some of the most salient issues concerning the relationship between migration, return, and development. It outlines the operational framework and research strategy that will be used to investigate this relationship in an ongoing research project focused on West Africa3. The paper first presents an overview of the contemporary trends in international migration, or more precisely intercontinental migration, regarding this geographical region. It then discusses the consequences that such migrations can have on domestic labour markets, the effects of migrant remittances, and the brain drain phenomenon. The return of migrants is addressed in a separate section because this part of the migration process deserves special attention if a better understanding of the migration-development linkage is sought. The developmental impacts of potential capital transfers occurring with return, and particularly the transfer of financial, human, and social capital, are discussed in the subsequent section. Finally, the last section of this paper presents elements for a meso-level approach to study these issues in order to achieve an improved understanding of the highly complex relationship between international migration, return, and development.

3 More details on this project, which covers Côte d’Ivoire and Ghana, are provided in section 6.

6

2. Contemporary trends in migration: the case of West Africa Migration has always been a key feature of people’s survival and advancement strategies in Africa. It has often been observed that there are few other regions in the world where the population is so mobile (Ricca, 1989). The highest concentration of migrants and refugees is registered in this part of the world, including a growing proportion of women and children (Adepoju, 1998; Russell et al., 1990). More than half of these migrants and refugees are to be found in West Africa, a region that is well known for its extensive migration systems which reach beyond the African continent 4.

Internal migration accounts for most migratory movements in West Africa (Adepoju, 1998)5. International migration remains predominantly intra-regional and occurs mainly between neighbouring countries due to ‘artificial boundaries demarcating socially homogeneous units into separate states’ (Adepoju, 1991b: 45)6. However, intercontinental migration to economically more advanced countries in the north has grown in importance due to the deteriorating socio-economic conditions affecting the region since the 1960s.

Intercontinental migration is directed to Europe, North America and the Gulf. These flows remain heavily male dominated, although more women today migrate over longer distances and for longer periods (Adepoju, 1991; 1995b; Findley and Williams, 1990). It is possible to distinguish between ‘forced’ and ‘voluntary’ migration flows7.

4 Although the quantity and quality of data on migration in sub-Saharan Africa have started to improve, there still is a considerable scarcity of reliable migration data. It is therefore difficult to determine the volume and the patterns of migration (Adepoju, 1995b; Foote et al. 1993; Zachariah and Condé 1981). 5 Recent literature on internal and international migration includes Baker and Aina (1995) and Cordell et al. (1996). 6 For studies on intra-regional migration of workers see Arthur (1991) and Brydon and Gould (1984). 7 It needs to be emphasised that migration is rarely a truly ‘voluntary’ activity. Certain people lacking resources such as money, information, and connections may not even have the choice to move. It therefore seems helpful to use the continuum between ‘forced’ and ‘voluntary’ migration as an analytical tool to establish different degrees of choice or freedom available to potential migrants, provided that the terms

The former include refugees and asylum seekers, whereas the latter can be classified into two main groups. On the one hand, there are the poorer unskilled migrants who tend to be pushed out by the adverse socio-economic conditions in the country of origin8. On the other hand, there are the highly skilled and educated migrants who usually come from the ranks and files of the urban elite. These migrants are generally pulled out by the greater training and work opportunities existing in the wealthier countries in the north9.

The direction of international flows is heavily influenced by historical and cultural ties. Migration occurs primarily between former colonies and past colonial powers because of their linguistic, administrative, political, investment, and transportation links. Migrants from French-speaking countries move mainly to France or Canada and those from English-speaking countries migrate principally to the United Kingdom or the Unites States. In recent times, however, a growing number of elite migrants from the former countries opt for the United States to pursue their studies as they are keen on diversifying their post graduate training.

The contemporary causes of emigration in West Africa are multiple and complex, including overall deteriorating socio-economic conditions, declining standards of living, and spreading poverty (Adepoju, 1995)10. A series of demographic,

‘refugee’ and ‘migrant’ are used exclusively (Faist, 2000: 23). 8 The great majority of migrants in the region are unskilled labourers (Zachariah 1980; Zachariah and Nair 1980). A good example of predominantly poor, unskilled migrants are the Soninké people who originate from the Senegal River Valley situated on the borders of Senegal, Mali, and Mauritania. The Soninké are well known for their circular migration strategies which reach far beyond the West African region. Although they come from very remote areas and many are illiterate, their preferred destination since the 1950s is France. Census data shows that 62,640 migrants from the Senegal River Valley lived in metropolitan France by 1982 (Findley and Sow, 1998). 9 African countries with relatively depressed economies such as Ghana, Guinea, Mozambique and Zambia tend to register greater losses of highly skilled migrants than countries with healthier economies (Zachariah and Condé 1981). 10 Economic reforms and structural adjustment programmes promoted during the 1980s and 1990s have been seen as an important determinant of emigration because they have enhanced unemployment in both public and private sectors (Adepoju, 1995).

7

economic, social, political, and even cultural factors determine the volume and the composition of intercontinental migration flows. A rapid growth in the population and labour force, sharp rises in un- and underemployment, decreasing real incomes and growing costs of living, low agricultural production and ecological disasters determining food shortages, and political instability and crises are only a few of the more generally accepted causes of out-migration.

The decision to migrate is rarely made by individuals acting on their own. Rather, migration decision-making often involves entire families as well as wider social structures and networks. Cordell et al. (1997: 15) have stressed that in West Africa migration results from a complex series of implicit and explicit ‘negotiations’ occurring within the household. This has also been confirmed by other studies emphasising the role of traditional hierarchies and gerontocracy in the decision on who should move, where, when, and the mode of behaviour at destination (Findley and Sow, 1998)11. Adepoju (1995a: 329) describes this decision-making process as follows:

’In maximizing household resource allocation and utilization, senior members of the family decide who should migrate just as they decide what piece of land should be cultivated and who should be sent to school. Households normally select and invest in a migrant who has the greatest potential for supporting the entire household in terms of remittances, especially where migration is the survival strategy type (as in the Sahel) rather than the mobility type. With economic distress in the region, a large portion of migration is of the former type.’

It is clear that usually migration does not only concern those who go, but also those who stay (Hammar et al., 1997). But social ties between emigrants and their families and relations back home are particularly strong in the African context (Russell et al., 1990)12. Migrant families and relations bear part or all the costs of migration. In return they expect migrants to send home remittances and gifts, to visit regularly, and to

11 See also Makinwa- Adebusoye (1995) and Oucho (1994). 12 This is one of the most consistent findings of the extensive research conducted by Russell et al. (1990) on international migration in sub-Saharan Africa using secondary and primary sources of information.

remain in contact by mail or phone. Migrants may also be encouraged to marry locally to guarantee that eventually they will return home (Findley and Sow, 1995).

Migration is an integral component of many African families’ survival and mobility strategies, which often culminate in return (Oucho, 1994). Decisions to stay, to leave, or indeed to return are closely related to the nature of the household economy and domestic and social power structures. But personal abilities and ambitions as well as structural and institutional factors operating at the local, regional, and global levels also play a highly critical role.

3. Consequences of international migration The issue of the consequences of international migration is of course intimately related to that of the interrelations between international migration and development. Migration effects can be examined at three main levels: the individual level, the family or household and local community level, and the wider national level. Varying effects are generally registered at the micro, meso and macro levels. For example, emigration of the highly skilled may have direct positive returns for individual migrants and their families (e.g. better standards of living and remittances), but may concurrently cause adverse effects for the national economy (e.g. loss of skilled manpower).

The consequences of migration also vary considerably according to the time perspective that is chosen. Certain consequences that appear positive in the short run, may turn out to be negative in the medium or long run. For instance, Fischer et al. (1997) have stressed that migration eases labour market and balance of payment problems and sometimes produces some growth effects to greater consumption determined by remittance flows. They have argued that in the short term migration increases wages in the emigration country and decreases them in the immigration country. It thus can be argued that migration has a negative effect on workers' living conditions in this latter country. In the medium term, however, migration may actually produce wage increases in both emigration and immigration countries because convergence rather than divergence effects prevail.

8

Most important, however, the consequences of migration differ depending on its volume and type. The consequences are obviously quite different when only few persons migrate or in the case of mass migration. But migration effects also vary considerably according to different types of migration such as internal and international migration, temporary and permanent migration, regular and illegal migration13.

The characteristics of migrants and the degree and direction of selectivity are also important determinants of variations in migration impacts. Migration of the highly skilled does not have the same effects than migration of the unskilled. But there are differentials in migration impacts not only by education but also by age and sex. Balán (1988) stresses that ‘migrant selectivity tells us much about the causal process involved in migration, since differential probabilities of migration involved in selectivity are usually linked to the social structure in areas of origin and the specific opportunities open through migration in the areas of destination’. However, migrant selectivity also provides important clues to determine migration consequences in both sending and receiving areas.

Important variations in the consequences of migration are also registered depending on the socio-economic and cultural situation in the areas or countries involved in the migration. These multiple factors influencing the nature of migration impacts are rarely taken into due consideration.

The review of the literature shows that the consequences of migration for the sending countries have generally been assessed at three levels: the effects of the departure of workers on labour markets; the impact of remittances on the national economy and the living standards of migrants’ family members; the implications of the loss of highly skilled migrants on the availability of human capacity necessary to harness development. Diverging views and contradictory findings concerning these issues will be discussed in the subsequen0t sections.

13 Migration types can be distinguished using spatial, temporal, administrative, and legal criteria. See for example Boyle et al. (1998) and Lewis (1982).

3.1 Labour market effects

There always has been a heated debate regarding the effects of emigration on labour markets in sending countries. This issue is obviously of particular importance in the West African context characterised by rapid population growth and increasing unemployment. Recent demographic forecasts suggest that the regional labour force in the year 2025 (477.30 million) will nearly match the region’s population in 1990 (482.53 million) (Adepoju, 1998: 302).

Neoclassical economic theory expects out-migration to bring about a positive impact on labour markets under such pressure, by easing un- and underemployment14. In these countries labour is generally abundant and capital is relatively scarce. Wages are lower here than in the more advanced industrial countries where the situation is the opposite. The higher wages existing in the industrial countries attract labour migrants from the developing world. International migration, deriving from these imbalances, results in increased labour supply in immigration countries and decreased labour demand in emigration countries. As a result, wage levels rise in the emigration country and drop in the immigration country until an equilibrium is reached. At that point, one might expect return migration to be initiated15.

However, some studies have shown that emigration can itself produce labour shortages in certain sectors of the economy. Adepoju (1991b) reports that in sub-Saharan Africa this has occurred mainly in agriculture and stresses that development may be obstructed as the young and enterprising adult males migrate while the older and more conservative stay home. Others, though, have demonstrated that labour shortages resulting from emigration can induce technological change encompassing a more rational use of resources. For example, Findley and Sow's (1998) longitudinal study on rural emigration in Mali shows that manpower 14 See for example Lewis (1954), Ranis and Fei (1961), Harris and Todaro (1970), Todaro (1976). 15

This model has been criticised because the decision to migrate is not only influenced by economic factors. It is also determined by social, political, and even cultural factors. It also has been stressed that its underlying assumptions, such as the homogeneity of labour as a production factor and the conviction that markets normally reach a stable equilibrium, are too rigid (Fischer et al., 1997).

9

shortages have produced favourable changes in farming practices. Households with migrants in France had to take on outside labour to replace absent labour. This strategy has had positive effects because it has allowed them to extend their cultivation by using part of the production to pay the labourers in kind.

3.2 Migrant remittances

It is often emphasised that while emigration countries lose manpower – and particularly the ‘best and brightest’ in the words of Papademetriou and Martin (1991) – they also get something in return. Migrants who work abroad usually send part of their income to their family in the home country. The economic impact of these remittances has been considered beneficial at both the micro and macro levels at least in the short term. Remittances can be an important source of income for families and households, and the importance of the total volume of remittances is undeniable16.

Hermele (1997) compiled fairly recent data showing that the aggregate value of migrant remittances can significantly exceed that of national export earnings17. For certain countries remittances are a more important source of revenue than official development assistance (ODA) disbursements (Livi-Bacci, 1993)18. The drawback of such high monetary transfers, however, is that developing countries may easily become dependent earnings that are nonetheless uncertain and vulnerable due to changes in migration policies and economic or political crisis (Appleyard, 1989)19.

16 Russell (1992) has stressed the problems related to measuring migrant remittances because of problems of definition. See also Russell (1986). 17 The total value of migrant remittances in 1995 is estimated at US$70 billion worldwide in 1995 by Taylor (1999). 18 World Bank estimates illustrate for example that in 1988 Egypt received $3.3 million or 10% of GNP in remittances. The same year ODA disbursements to Egypt were only 45% of that value. Meanwhile, recent work by Ahmed (2000) in Somaliland has estimated remittances at US$500 million annually, around four times the value of the main export, livestock. 19 For example, significant remittance flows were interrupted during the Gulf war (1990-1) because it produced sudden returns of migrants to their home countries (Skeldon, 1997).

The amounts that migrants remit vary depending on their income, propensity to save, and length of stay. Institutional arrangements for money transfers, exchange rates and risk factors, distance of the immigration area, and strength of social ties also influence the amount of remittances (Adepoju, 1991b; Taylor, 1999)20. Remittances increase proportionally with the income of migrants, while they decrease with the length of stay abroad (King, 1986). Remittance money is likely to be more important when migrants plan to return than if they intend to settle for good in the host country.

The amounts remitted also tend to increase with the distance and the costs of migration. This seems to confirm that migrant families expect a monetary return to the investment in migration of one or more family members who have been sent abroad. Studies conducted in West Africa confirm that remittances are part of an ‘implicit contractual arrangement’ between emigrants and their families. Russell et al. (1990: 33) reviewed various empirical migration studies in sub-Saharan Africa showing that remittances increased in periods of crisis as they were an integral component of risk diversification strategies pursued by families to ensure support in moments of need. A positive correlation was also found between amounts remitted and per capita household income from other sources.

The impact of remittances in the long term has been the focus of considerable debate. Many scholars have argued that remittances fail to enhance development because they are not spent on investment goods but rather on basic consumer goods. Hermele (1997: 136) describes how this argument goes: ‘Remittances are not put to productive use, but mostly spent for unproductive purposes - housing, land purchase, transport, repayment of debt – or, to a smaller degree, wasted on conspicuous consumption, or simply saved as insurance and old-age pension funds’. However, this point hinges on the definition of ‘productive’ and ‘unproductive’ uses of remittances. Some ‘wasteful expenditures’ on food, shelter, land, and especially education may

20

Various channels are used to sent or bring back remittances in the form of money or consumer and durable goods. Ghosh (1992) has stressed the importance of promoting the use of regular channels because the use of irregular ones may be directly or indirectly related to immoral or illegal activities such as the exploitation of poor and uneducated migrants.

10

have a positive effect in the long run. For example, Taylor (1999) points out that building a house has an impact on family health and village construction activities, whilst Lowell and Findlay (2001) note the ‘induced effect’ of emigration on increasing demand for education, and private funds to support it, may have a highly beneficial developmental effect.

Other scholars have emphasised other problems. For example, Piore (1979) has argued that remittances cause inflation because they create a demand without concomitant production capacity21, whilst Böhning (1984) emphasised that remittances increase the demand for imported goods, producing a negative effect on the balance of payments22. Social inequalities have been seen as another negative effect of remittances. Lipton (1980, 1982), for instance, has argued that remittances sharpen income inequality because the better-off parts of communities are more likely to send migrants abroad and thus are also more likely to draw greater benefits from migration (Ammassari, 1994).

On the other hand, it has been stressed that remittances have a positive impact on the balance of payments of sending countries as they help to narrow the trade gap, control external debt, facilitate debt servicing, and produce much needed foreign exchange (Appleyard, 1989). Empirical evidence from different countries also contradicts pessimistic views regarding the use of remittances. From their study on sub-Saharan Africa, Russell et al. (1990) concluded that ‘once subsistence needs are met, migrants do use remittances for investment purposes including education, livestock, farming, and small scale enterprise’23. Findley and Sow (1998) report from Mali that remittances not only covered basic food and cash needs but also allowed to pay for irrigation in agriculture. Remittances have also been a significant source of financing for 21 Thomas-Hope’s (1999) findings from Jamaica confirm this view showing that migrants were charged higher prices for houses and services. 22 Certain scholars set emphasis on the ‘remittance multiplier effect’ arguing that expenditures of remittance-receiving households may create a demonstration effect on households that do not receive any (Appleyard, 1989). 23 It is important to notice that there are differences in the use of remittance by sex (Russell et al., 1990) and educational status (Athukurala, 1990). Palmer (1985) has focused on the role of women and household structure and remittance patterns in Lesotho.

agricultural inputs in Zambia (Chilivumbu 1985). Recent work in Somaliland has highlighted investment of remittances in production even in highly unfavourable economic and political conditions (Ahmed, 2000). Meanwhile, Gustafsson and Makonnen (1994) have found for Lesotho that migrant remittances actually decrease inequality.

In a much earlier study of internal migration in Ghana, Caldwell (1969) found that migrants spent remittances to pay for schooling and wages of farm labourers, and to develop small businesses. In other instances remittances have been used to finance social and economic development projects in the home country. In their recent study on Senegal, Diatta and Mbow (1999: 246), found that remittances were a substantial source of revenue for migrant families and were also used to promote development in migrants’ home communities.

3.3 The brain drain

Migration of the educated and skilled from sub-Saharan Africa countries to industrial countries in the North has reached significant proportions, and there is little evidence that these flows will decrease in the near future (Appleyard, 1998; Adepoju, 1995). These prospects inevitably raise deep concerns among policy makers who continue to call for effective policy measures that can facilitate the return of migrants or to encourage them to maintain links with their home country (Grey-Johnson, 1986; Ndioro Ndiaye, 2000; Pires, 1992).

It is estimated that today about one third of the most highly qualified African nationals live outside their country of origin, mainly in Western Europe and North America (World Bank, 2000). Logan (1992) published data on US immigration showing that a larger proportion of Africa’s total migrant pool is made up of professionals compared to the rest of the world and that the rate of growth of the professional migrant flow from Africa is generally higher. UNCTAD estimates that, between 1960 and 1987, about 70,000 highly skilled Africans left the continent to access the West (quoted in Russell et al., 1990). This corresponds to approximately 30% of the high-level manpower stock available within the continent at that time, an estimate confirmed this year by the World Bank (2000: 39).

11

The brain drain has also been the subject of highly controversial debates since the 1960s24. The divergence theory sets emphasis on the highly detrimental effects of these flows claiming that developing countries are losing their best people from key positions, and thereby weakening human capacity to harness modern agricultural and industrial technology25. The emigration country looses critical human capital in which it has invested through education and training and for which it is not compensated26.

On the other hand, representatives of the convergence school argue that the problem of the brain drain is less critical than it is usually portrayed. They tend to emphasise the beneficial consequences of migration for the receiving and sending countries and especially the positive effects of emigration on (un)employment27. It is argued that if highly educated and skilled people were not to emigrate they would be under-utilised at home due to unemployment or poor working conditions which may suffocate their inventive and creative abilities that can better be used abroad (Keely, 1986). It is also stressed that human capital investments made in the high-level migrants are partly recovered through remittances.

Although the brain drain dilemma remains largely unresolved, some suggestions to solve the

24 See Bennell and Godfrey (1980), Chukunta (1979), Das (1979) and Gardiner (1968) for a special focus on Africa; Glaser (1978) and Bhagwati (1976) for more general overviews. 25 It is implicitly assumed that the highly skilled have a greater propensity to move, because for them the benefits of migration tend to be higher due to a higher demand and smaller costs of mobility (Fischer et al., 1997; Skeldon, 1997). 26 A major aspect of the brain drain controversy is whether high-level skills are drawn off unfairly. 27 A survey conducted in 1985 in Côte d’Ivoire shows that 37.5% of people with a university diploma in Abidjan, and more than 50% of such people in other cities, were economically inactive or unemployed (Russell et al., 1992). More than half of these people had been in that situation continually for the year preceding the survey. Among those people with vocational and technical diplomas, 33% were lacking employment in Abidjan, and 26% in other cities. The results also showed that about 10% of graduates had started working in the informal sector as self-employed entrepreneurs.

problem have been put forward (Obia, 1993)28. Concrete policy initiatives have also been devised to foster the return of highly educated and skilled migrants and to encourage their support in the development of their home country (Ghosh, 2000b)29. ‘Return of talent’ programmes have been implemented in various countries in Asia, Latin America, sub-Saharan Africa, and the Caribbean. Some have been sponsored directly by Governments, others through multilateral organisations30.

4. The study of return and transnational migration: methodological issues The return of migrants with its human capital implications is one of the most commonly cited benefits of migration for the sending country, along with remittance and labour market effects (Todaro, 1976). But this is not the only form of capital transfer involved with return migration. Financial and social capital can also be mobilised through migrants' return. Migrants may accumulate savings while living and working abroad and bring them back once they return. They may also make professional and personal contacts, which prove useful and productive for their endeavours back home. It is argued that these implications of return migration vary significantly according to the level of analysis that is chosen and based on several critical factors. These include the volume of return migration, characteristics of migrants, degree and direction of selectivity, types of migration, reasons for return, and situations existing in the countries involved in the migration.

Regardless of its importance for the understanding of the migration-development linkage, return migration has always been heavily

28 See also Ghosh (1992), Dumon (1984) and Condé (1989) for policy measures concerning return migration more generally. 29 Not only has the support of highly skilled migrants been sought but also that of unskilled migrants who can act as ‘partners in development co-operation’ (Libercier and Schneider, 1996). 30 See also Koser (2000) on different types of policy frameworks and operational programmes, Logan (1990) and Ardittis (1985) on UNDP-sponsored TOKTEN Programme; Pires (1992) and Ardittis (1991) on IOM’s Return of Qualified Nationals Programme, and Strachnan (1980) and Abraham (1968) for examples of return supported by government.

12

disregarded in the migration literature (King, 2000). Even a superficial review of this literature shows that very few articles and books deal appropriately with this critical issue. But while empirical studies on return migration have started to accumulate, theoretical approaches and models to study return migration and its development implications are still lacking.

Interest in return migration first became apparent in the mid-1970s, following the economic crisis of 1973. Various measures needed to be devised to enhance the European Gastarbeiter’s return to their countries of origin and facilitate their reintegration back home. In fact, the relatively short history of return migration can be traced in three main generations of migration studies. The first generation focused on the push and pull factors of migration in the context of industrialisation. Flows were mostly seen as distinct movements from emigration to immigration areas, but the existence of ‘counter-streams’, including migrants who return home, was acknowledged. Lee (1964:54-56), for instance, argued that some migrants return because they have acquired new attributes at destination, which promise to be advantageous at origin (e.g. skills or wealth).

The second generation of migration studies emphasised the nature of structured relationships between the emigration areas (periphery) and the immigration areas (core) (see for example Portes and Walton, 1981; Mabogunje, 1970). Both areas were seen as tied into a general migration system. Not only migration flows linked these areas together, but also flows of information, capital, goods and services. Historical and cultural linkages facilitated these flows and nurture the reciprocal relationships. In extensions of this perspective, return migration was considered as a form of feedback from the area of destination to the area of origin. Social network theory, for example, has posited that information provided by returnees and their post-return behaviour may either stimulate or deter further emigration (Boyd, 1989; Fawcett, 1989).

The third generation of studies on contemporary international migration is currently developing. The new conceptual framework suggests that in the context of globalisation areas of origin and of destination are linked together in transnational spaces by migrants who build and maintain simultaneous and multi-stranded relations that span boundaries (Glick-Schiller et al., 1995; Smith

and Guarnizo, 1998). In this view, return migration is a central part of the ongoing migration processes. It is no longer seen as the ‘closure’ of the migration cycle, but rather as one of the multiple steps of a continued movement (King, 2000).

4.1 Problems of definition and measurement

Defining return migration and other closely related concepts in the present context characterised by accelerating and increasingly multifaceted migration processes is a difficult task. But it is of course necessary to clarify some of these concepts to allow for measurements and comparisons within the framework of an empirical study31. King (2000: 8) has elaborated on the early work of Bovenkerk (1974) to define return migration as a ‘process whereby people return to their country or place of origin after a significant period in another country or region’. However, although useful, this definition suffers from the problem that it is not clear what is meant by 'significant period'. Indeed, the time dimension is an essential aspect that needs to be clarified in operational terms if an empirical investigation is to be conducted on the ground.

No measurement is possible unless the time dimension of migration and return is adequately specified. This means that the length of stay

31 The need to elucidate the concept of return migration to allow for a reliable measurement of this phenomenon and sound understanding of its implications has long been stressed by the UN Secretariat (1986). It has also been suggested that a universal definition be adopted in different contexts of study to allow for comparisons. The UN Recommendations on Statistics of International Migration suggest that the category of ‘short-term emigrants returning’ includes ‘persons (excluding border workers) who have entered the country, who had previously been in the country at least once for more than a year and not away continuously for more than one year, and whose last departure was to work abroad at an occupation remunerated from a foreign country or to accompany or join such a person as a dependant or domestic employee’. Return migrants who have been abroad for more than a year would figure among the ‘long-term immigrants’ provided that they intend to stay in their home country for at least a year. Return migrants who plan to stay for a shorter period or do not intend to work for a year would be included into the category ‘other immigrants’. It is clear that the application of this definition does not easily allow identification of return migrants.

13

abroad and the intended length of stay back home need to be determined. Yet there is a tension between adopting a broad timeframe including also very short stays abroad to adequately capture transnational migration, and still being able to distinguish between simple visits and migration. There are also other key aspects that call for clarification. The reasons of departure and return need to be clarified not only to keep 'visitors' separate, but also to avoid confusion between people who go abroad to work and those who leave to accompany or join these persons. Another category that may be considered separately is the one comprising persons who go abroad to study.

Two other criteria that seem of particular importance are nationality and place of birth. While the latter is straight forward, the former can be a source of some confusion. Usually only persons possessing the nationality of the country they are returning to are considered ‘return migrants’. There may however be cases of migrants having given up their nationality when taking up a new one. There are also persons having the nationality of the country they are entering, but who are born and have lived all their lives abroad – is it possible for second or third generation Ghanaians, born in the UK, to ‘return’ to Ghana?

It is also useful to distinguish highly skilled and unskilled migrants, given the potentially different impacts of these groups on development noted in the previous section. Yet, it is clear that hard and fast divisions between the two are difficult to sustain. Rather than talking of 'highly skilled migrants', it is perhaps better to speak of those who in their country of origin belong to the 'elite'. This represents a minority, a group of people of relatively high status in terms of education, occupation, wealth and power32. When these persons return they often access top positions of responsibility and authority and therefore have a fair ability to influence the course of events in

32 In the Ghanaian and Ivorian context eliteness seems to comprise a combination of some or all of the following: (i) family background, (ii) high levels of education, which could have been acquired home or abroad, or a mixture of both, (iii) exercise of power, through occupation/employment, class or whatever, (iv) wealth, and perhaps, finally (v) a self-perception as being part of an elite – a sort of ‘elite consciousness’.

different spheres of society33. 'Unskilled migrants', instead, comprise those persons who have little or no formal education. They usually come from the lower strata of society, but may of course become socially mobile through migration.

4.2 Characteristics of migrants and multiple selectivity

The existing body of empirical evidence confirms that migration is a selective process. Migrants are not a random sample of the population living in the country of origin; rather, young adult males are generally thought more likely to migrate, as are single persons – although there is some evidence that the movement of single women migrants has been under-emphasised (Brydon, 1979; 1992). Schooling increases the propensity to migrate, and so does wealth as it provides the resources to move. There seems to be much consensus that the better-off and more educated migrants are more likely to migrate more frequently and for longer distances (Lewis, 1982)34. Yet, as an apparent paradox, high propensities to migrate also tend to be associated also with low education and illiteracy (Lipton, 1982).

Given the cumulative personal and contextual aspects that characterise the multidimensional selectivity of migration, it is possible to distinguish between two main categories of migrants. On the one hand, there are the poorer (though not poorest) unskilled people for whom migration represents a survival strategy. This group of people often have little choice other than to migrate. On the other hand, there are the wealthier and highly skilled people who choose migration as a strategy for advancement. For them migration represents one possible way to improve their own and family status and to be socially mobile.

Lipton (1982: 20), who focused on rural to urban migration, has emphasised this positive and

33 ‘Highly skilled return migrants’ include for example senior government officials, chief executive officers, managers, senior professionals in private companies, owners of firms, professors, political party leaders, doctors, lawyers, judges, and heads of NGOs. 34 Rose (1970) studied the relationship between the number of migrants and distance and concluded that lower class people migrate for shorter distances as they find many more intervening opportunities in a given distance than upper class people.

14

negative selectivity of migration. He has argued that ‘if we ranked villagers into ten - sized groups by income-per-person, migrants would probably be likeliest to come from the second-poorest and third-poorest - and from the second-richest and third-richest groups’. The reason is that the poorest do not have the choice to migrate as they lack resources, and the richest are already in such a good position that migration may become superfluous.

It is generally recognised that migration selectivity deserves a central focus in studies concerned with the impact of migration both in the areas of destination and the areas of origin (Lewis, 1982). First, there generally are important demographic consequences, such as delayed family formation and related effects on fertility rates. Second, age and sex selectivity may produce changes in the division of labour as well as residential patterns and household arrangements. Third, the impact of selectivity by socio-economic status on community structure depends on whether migration occurs with or without return as the latter drains local resources and constitutes a source of change in the distribution of wealth and political control (Balán, 1988).

If emigration is a selective process, then so is return. However, both the nature of selectivity and its impacts are yet to be explored in depth. Some research has focused on the question whether return migrants are rather 'successes' or 'failures' (Bovenkerk, 1974; Gmelch, 1980), but there are contradictory findings, whilst the question itself is problematic (King, 1986a). Another problem is that migrant selectivity has been measured in different ways. Some studies compare return migrants with non-migrants in the area of origin, whereas others compare return migrants with emigrants who remain in the areas of destination. Regarding this issue, King (2000: 42) draws the following conclusion:

‘Insofar as some generalization is possible, the return flow contains a disproportionate number of migrants with ‘low personal effectiveness’, and of migrants who are single, separated and divorced. On the other hand, some recent studies note that returnees tend to have higher levels of education, skill and income than non-returnees, particularly where opportunities for professional advancement exist in the home country.’

For Gmelch (1980) return migrants are neither great successes, nor particular failures. The highly successful emigrants are reluctant to go back because they do not want to quit their secure and well-paid jobs. They are also concerned about costly obligations vis-à-vis family and kin who expect them to share the benefits of their success. The very unsuccessful emigrants also rarely return because they do not have enough resources to return and more importantly they do not want to admit their failure.

Similarly, from his extensive literature review Bovenkerk (1974: 23) concluded that ‘the returnees of permanent emigration were seldom failures, the returnees of temporary migrations were seldom successful. In the case of permanent emigration this can be explained in two ways: firstly, the total emigration stream contains a subsidiary flow of intending temporary migrants…. Secondly, it is possible that emigrant failures have not enough money to return and/or are reluctant to demonstrate their obvious failure back home’. These hypotheses need to be considered in the light of the migrants' characteristics. Some recent migration research suggests that highly skilled migrants not only are more likely to return, but also more likely to act as 'agents of change' (Thomas-Hope, 1999).

4.3 Types of migration and return

As there are different forms of migration, there also are varying types of return. These differentials are important factors influencing the impact of migration and return on development. A very common distinction is the one between temporary and permanent returns, which is based on the intention of the returning migrant to go abroad again or to settle for good back home. Two more categories can be established on the basis of the temporal criterion: occasional returns and seasonal returns. The first consist in short-term visits to see the family or to conduct business for example. The second involve temporary stays, which are determined by the seasonal character of the activities conducted, as for example agricultural work in the home country that requires significant labour inputs35.

35 Focussing on skilled migrants, Gould (1988) has classified circulatory flows into contract workers moving between developing countries, trainees and self-employed professionals migrating from developing to economically more advanced countries, personnel of multinational corporations moving between more

15

Another classification that is important to discern the consequences of return is the one conceived by Bovenkerk (1974) and later refined by Gmelch (1980) and King et al. (1983, 2000). This is based on two main criteria: the lengths of time that migrants intended to stay abroad and their return orientation. It encompasses four categories: (1) intended temporary migration with return; (2) intended temporary migration without return; (3) intended permanent emigration with return; and (4) intended permanent emigration without return36. This classification is useful as empirical studies have shown that capital transfers in the form of remittances, for example, are more significant when migrants intend to return home.

However, typologies using the intentions of migrants as criteria for classification usually present some problems. Firstly, migrants do not necessarily have definite plans at the time of their departure to go abroad or to return (Gmelch, 1980). Secondly, migrants' intentions may change over time depending also on opportunities and constraints arising abroad and back home, their own and their family members’ lifecycle stage, and various professional or personal motives. Indeed, it has been argued that the intentions of migrants are a poor indicator of actual migration behaviour37.

In contrast, King (2000: 9-10) has suggested another typology of return migration, which is based on the level of development of the countries involved in the migration. Three forms of return are contemplated in this respect: (1) return from developing to industrial countries that were linked by colonial ties (e.g., the British from India or Kenya, the French from Algeria); (2)

advanced countries or between these latter and developing countries. Later work by Findlay and Salt (1989) and Findlay (1990) has helped elaborating this typology and analysing migration implications by the type of movement. 36 Anwar (1979) has introduced the discussion about ‘the myth of return’, a term that describes the contradictory feelings that migrants nurture via-à-vis their home country and the country where they live. A return remains always envisaged, no matter how settled migrants are in the host country. But the return is constantly postponed although remittances are sent regularly and tight contacts are maintained with home. 37 See also Bovenkerk (1974: 18) who argues that ‘migrants do not always have a clear idea about their intentions and, moreover, tend to rationalize their reasons into motives that are accepted by the community as legitimate standard-motives’.

return from countries who have achieved a relatively similar economic development level (e.g., Belgians from France); and (3) return from developing to industrial countries whether linked by colonial ties or not (e.g., the Ghanaians from Britain or the US, the Ivorians from France or Canada). It is the absence of this last type of return of the highly skilled which is the subject of much concern and controversy.

The most elaborate typology of return migration is the one that has been developed by Cerase (1974) who studied the return of migrants from the United States to rural areas in southern Italy38. Focussing on the length of time the emigrants spent abroad and the degree of their acculturation in the host society, he has developed a typology including four different categories of return. The return of failure includes migrants who have failed to overcome the ‘traumatic shock’ upon arrival and who were not able to adapt to the new environment. These ‘failures’ represent only a minority and re-adapted quite easily to their home context because they return quickly before their integration in the host society starts. The return of conservatism involves those migrants who have migrated to pursue a specific objective and thus save a significant portion of their income. These migrants stay longer abroad and during their absence transfer remittances or save some money to realise their plans back home. This means that the values of the home society continue to prevail as a reference for measuring success and achievement. The return of motivation concerns migrants who stay in the host society long enough to start referring to its value system, but who eventually return home. These migrants are the ones who bring back new ideas and values and are ready to apply the skills they have acquired abroad in their home context. The return of retirement regards migrants who have terminated their working lives and go back home to retire. The third type is the form of return that has 'positive' implications on the sending country. Cerase is however quite sceptical that returnees can successfully assume the role of ‘agents of change’. He argues that this is only possible if

38 It is important to note that the great majority of the returned migrants that Cerase interviewed originated in rural communities, but settled and lived during their period overseas in an urban area. This means that they ‘experienced an abrupt passage from a rural world to an urban and industrial one’ (Cerase, 1974: 249).

16

their actions can affect the economic structure and power relations of the home communities.

4.4 Return motives and readjustment problems

The influence of return migration on socio-economic change in the emigration country also varies depending on the motives for return and the problems that returnees encounter. Return motives include a variety of economic, social and family-related, as well as political reasons (King, 2000)39. Existing empirical evidence shows that social and family-related reasons are of particular importance. Strong family ties, the wish to rejoin family and friends, homesickness, problems of adjustment in the host country, racial harassment, and the aim to enjoy an improved social status back home are significant reasons for return. Other factors are related to migrants’ stage in the life-cycle, as age brings changing needs and preferences. Migrants may for instance wish to raise their children back in their home country. They may return to get married, to care for elderly parents, or to take on particular family-related responsibilities.

Some migration scholars have examined the relationship between integration and assimilation and return (Hoffmann-Nowotny, 1973; Esser, 1980). They have claimed that return occurred because migrants were unable to integrate or assimilate into the host society. Rogers (1984: 233) has argued that ’Returns may indeed indicate failure to become integrated and assimilated into the host country, but they may not do so. For some migrants they may be ‘new’ decisions, arrived at on a basis of changed personal and societal circumstances.’40 Indeed, Al-Ali et al. (2001) show how increased engagement of Bosnian and Eritrean refugees in their home countries – including a willingness to contemplate return – can sometimes come more from security in their host state, rather than rejection by it.

39 See also Bovenkerk (1974: 20-25), Gmelch (1980: 138-41) and King et al. (1983: 25-31). 40 In her research on Yugoslavian migrants in Austria, Rogers (1984) found that migrants' attachments to home and host countries co-exist. The attachment of migrants to their host country was mainly instrumental, whereas the attachment to the home country was often sentimental.

There are also various studies of the difficulties that migrants face upon return (Bovenkerk, 1974; Gmelch, 1980; King, 2000; and Lepore, 1986). Gmelch (1980) has distinguished two perspectives from which this question may be approached. On the one hand, the actual economic and social situation of returnees can be examined, looking at employment and housing, participation in associations, and so on. On the other hand, the migrants' own perceptions can be measured based on the degree of 'satisfaction' or 'dissatisfaction' with regards to their expectations nurtured vis-à-vis the home country upon return.

Readjustment problems have been considered as an indicator of social change brought about by return migrants. Cerase (1974), for instance, has postulated that the greater the conflict of re-adaptation due to the changed value structure of the return migrants, the greater the probability that they will provoke social change. Conversely, the more traditional their value orientation upon return, the less likely they are to bring about innovation and social change. King (2000) has indicated that these hypotheses find confirmation in much existing empirical evidence. But he also has emphasised the ambivalence that accompanies migrants' return. On the one hand migrants are relieved to be back home in a familiar context. On the other hand, they are often quite disappointed after return and easily irritated with how things are done at home. Also, the relationship between returnees and non-migrants is generally ambiguous (Gmelch 1980).

5 Development implications of different forms of capital transfer The development implications of return migration have so far been explored along two major lines: financial capital in terms of savings accumulated by migrants while working in the host country, and human capital acquired abroad in the form of education, training, and working experience. But a third form of capital transfer also seems of critical importance: social capital transfers. It is essential to examine all three forms of capital transfer to gain an improved understanding of the relationship between migration and development.

A critical question in this respect is whether permanent return is a prerequisite for making transfers of financial, human, and social capital advantageous for the development of migrants' home countries. Or is there a potential in the

17

context of globalisation and transnationalism for migrants to contribute to such development without settling there for good. Indeed, some studies suggest that circulatory migration even at a sub-regional level can have a positive impact on investment, as Williams (1999) reports for improvements in livestock management in West Africa.

For Ghosh (2000b) two major factors determine the impact of return on personal success and development: (a) the aptitude of the returning migrant, and the degree of preparation of return, and (b) the socio-economic and institutional situation existing in the home country. However, to this we can add a number of other factors: the magnitude of migration flows, migrant selectivity, migration types and motives, and the characteristics of emigration and immigration countries.

The inconclusive nature of empirical findings on the developmental impacts of migration and return results in part from the fact that few studies distinguish impacts according to these critical variables (King 2000). This had already been emphasised by scholars in the mid-1970s. In his bibliographical essay, Bovenkerk (1974: 46) has stressed that empirical evidence concerning the relationship between return and innovation was ‘too little, too unreliable, too unspecific, too fragmentary and as said too partial, to make possible comparisons of the conditions under which innovation by return migrants succeeded and in which attempted innovation became failure’. He has set emphasis on the difficulties associated with weighing up the relative importance of various factors, and suggested nine factors that may influence the potential of migration and return as a development force (see section 6.2).

Bovenkerk (1974: 45) has also stressed the importance of defining what is meant by the terms innovation, change, and development. He reveals his disappointment with regards to the fact that

‘In almost all the literature in which the influence of returned migrants on the social and economic development has been studied, development and the various innovative stages that lead to it, is implicitly defined in terms of the model presented by the rich Western countries of the world. This is supposedly the viewpoint of most returnees

themselves, because they bring back influences from these very countries. There may, in these sources, be some discussions about the way Western ideas are to be translated by the returnees in the context of their home countries, but the usefulness of these ideas themselves are never questioned.’

This raises an important issue, which deserves attention at two levels: on the one hand, at that of scientific discourse and interpretative schemes, and at that of migrants' life history reports and self- definitions, on the other hand.

5.1 Financial capital

Some migrants who live and work abroad are able to save money, which they may transfer to the home country upon return. These transfers of savings should not be confused with those of remittances, which are the payments that migrants send back home to their families while they are still living abroad. A distinction is rarely made in empirical studies between these two concepts, which need instead to be kept separate (King, 1986; Russell et al., 1992).

The impact of migrants’ savings has raised a very similar controversy to the one concerning the effects of remittances. While some scholars have stressed the positive implications of migrants’ savings for migrants, their families and their countries, others have argued that the impact of savings generally is insignificant and sometimes even negative. Existing empirical evidence demonstrates that migrants transfer substantial amounts upon their return as they usually earn more than what they would earn at home and try to limit consumption. But the evidence concerning the use to which migrant savings are put is quite contradictory.

King (1986) indicates that the amount that migrants save depends on the level of their wage and necessary expenses for subsistence, accommodation, and transport. Many return migrants are very well off compared to non-migrants, and it seems clear that, ‘the greater the wealth and general well-being on return, the greater are likely to have been the sacrifice, self-denial and deprivation of the migrants during the 'other existence' abroad’ (King, 1986: 24)41.

41 A relatively vast body of evidence concerning social mobility of migrants as compared to non-migrants has

18

From their reviews of empirical studies on the use of migrant savings, Gmelch (1980) and King (1986) conclude that a fairly small portion of savings is spent for productive investments. Savings are largely spent on 'conspicuous consumption' to raise the status and comfort of return migrants and their family. A considerable proportion of savings goes for constructing a house or to purchase a plot of land. Migration scholars have often condemned this type of expenditure, arguing that migrants' savings and remittances could be invested elsewhere in a more productive way. Emphasis is also placed on the possibility that this form of spending may cause structural distortions such as inflation (Böhning, 1975).

Russell (1992), who has examined the uses and consequences of remittances, has argued that the distinction between consumption and production/investment expenditures is rather blurred. In fact, it can be argued that spending on housing, consumption, and services (e.g., education and health) may create employment and produce positive multiplier effects as well as reduce the need for government expenditure on infrastructure, subsidies, and services. Expenditures on housing may moreover serve not only to raise the status of return migrants and their family, but also provide them with a better access to other local resources.

5.2 Human capital

One of the most debated issues has been that of human capital gains for emigration countries through the return of migrants. It has often been claimed that migrants acquire valuable training and work experience when they study and/or work in more industrialised countries. New skills, ideas, and attitudes of returned migrants are expected to have a positive impact on the development of their home country. Much empirical evidence has however contradicted this optimistic view. Various studies from southern Europe, especially from the 1970s, have found that only a minority of migrants had gained new skills while working abroad (Gmelch, 1980; King, 1986). The majority did not learn anything new because they only got unskilled work to do. Work on migration to Europe from Africa and the Middle

been produced in the 1960s and 1970s. See for instance Dahya (1973); Gmelch (1979); Hernandez-Alvarez (1967; 1968); Krane (1973; 1976).

East has also identified deskilling of migrants as a major problem (Brydon, 1992; Al-Rasheed, 1992).

But even among those migrants' who are able to acquire new skills and experiences, few may be able to apply them in practice back home. It is obviously difficult for migrants who have acquired technical or industrial skills to apply them in rural settings lacking the infrastructure needed to make an effective use of their new skills (Castles and Kosack, 1973; Gmelch, 1980)42. Also, many migrants are unskilled and get to do unskilled work whilst abroad, which does not teach them anything new. Labour migrants often get very little training.

Böhning (1972), for example, studied the social and occupational apprenticeship of Mediterranean workers in West Germany and found very little evidence in support of the hypothesis that migrants return home equipped with new skills and the desire to operate change. Migrants had been largely engaged in low-grade positions in the industry and worked mainly on mass-production lines. They were often frustrated and this has hampered their learning. Abadan-Unat (1976) stressed that labour migrants with real industrial know-how tended to stay on in West Germany. Their expertise can hardly be used back home. Migrants who returned were those with very limited skills who tended to avoid industrial jobs back home. These migrants have learned very little beyond, for example, how to empty dustbins in Munich, turn a screw at Renault's, and wash dishes in Zürich.

But there also is empirical evidence gathered in other regions, which contradicts the conventional view of human capital theory considering labour market experience in industrial countries as an asset. Muschkin (1993), for example, has researched the consequences of return migrant status for employment in Puerto Rico. When she compared return migrants with non-migrants, she found that the former had higher unemployment rates and lower mean earnings. A longer duration of stay was moreover associated with negative employment outcomes. These findings have led

42 See also Piore (1979) who takes a critical standpoint stressing that returnees are not a significant source of industrial skills. The most skilled remain abroad whereas the failures tend to return. But returnees are not willing to go back to the agricultural work they did before leaving. This work is usually performed by immigrants from less developed countries.

19

her to postulate that the longer migrants were absent from home, the weaker their network ties, and the greater the difficulties of finding employment. But a greater length of absence also seemed to protract migrants' work in jobs with skill requisites that do not transfer well back home. On the contrary, studying the return of Pakistanis from the Middle East, Arif and Irfan (1997) have found that migrants had been able to move out of production-service occupations into business and agriculture occupations. This advancement was largely due to their experience abroad with the length of stay in the Middle East influencing occupational mobility.

There also is an important body of empirical evidence that seriously challenges empirical findings showing that return migrants produce only very little change. Returnees not only bring back new skills and capacities, but also innovative ideas and changed attitudes and behaviours (Gmelch, 1980). The question obviously is whether they have succeeded in introducing the foreign-acquired ideas and practices into their home communities.

In one of the earlier studies, Saloutos (1956) stressed the benefits of migration and return for Greece arguing that returnees brought back new ideas on democracy, health standards, and social behaviour. They introduced liberal business practices, and pro-US and pro-Western values. Additional empirical evidence collected in Greece much later has confirmed substantial differences existing in the attitudes and behaviours of returnees and non-migrants with regards to a series of social issues, such as divorce, dowries, gender relations, and child-rearing (Bernard, 1978).

The results of studies conducted in the Caribbean support the hypothesis that return migrants can act as agents of change under certain circumstances. In their study of a Barbadian village, Sutton and Makiesky (1975) found that returnees not only had acquired new values and beliefs, but also a greater racial and political consciousness. Comparing skilled and unskilled return migrants in Barbados, Gmelch (1987) found a significant proportion of student-migrants entering professional or white-collar occupations at home who were more likely to introduce new ideas and techniques acquired while abroad. Unskilled migrants had less impact because they went back to jobs that did not exploit their overseas work experience. They also lacked the

necessary positions of authority and power enabling them to induce change43.

Studies of return migration in Jamaica confirm that returnees can play a meaningful role in the development of the home country. Taylor (1976) reported that Jamaican migrants return home with accrued skills and experience. This has been confirmed by Thomas-Hope (1999) in a recent study showing that the majority of returnees from North America and Europe have gained additional skills and capacities as well as changed the way they were functioning in the workplace. Most important, however, returnees have an impact due to the leadership roles that they assume in their home communities44.

5.3 Social capital

While financial and human capital transfers have received attention in the migration literature, the transfer of social capital, consisting of a set of specific resources that can be mobilised within groups, networks and organisations, has largely been ignored. Social capital designates the potential wealth that can be drawn from social relations. Thus Bourdieu and Wacquant (1992: 11) define it as ‘the sum of the resources, actual or virtual, that accrue to an individual or a group by virtue of possessing a durable network of more or less institutionalized relationships of mutual acquaintance and recognition’. It is built on mutual obligations and expectations, norms of reciprocity, trust and solidarity.

A review of the literature shows that studies concerned with social capital have almost exclusively dealt with the extent of participation in networks and with the resources that derive from such participation. For the study of the transfer of social capital, however, another kind of resource assumes particular significance. This is the competence that people acquire in building and nurturing interpersonal relations and social ties in varied socio-economic, cultural and political

43 Gmelch (1980) has also examined the impact of return migrant outside the workplace and has stressed that it is hard to draw any meaningful conclusions as it is difficult to separate the role of return migrants as agents of change from other foreign influences such as films and television or tourism. 44 In this respect it is interesting to study the return of the elite who traditionally goes abroad to study (and, sometimes, to work) and the lives of leading individuals such as political leaders or successful businessmen.

20

contexts. Such competence is to be conceived as a subjective variable corresponding to the product of a formal and informal learning process that helps achieving ‘wider horizons’. Language skills, the ability to interact and work with people of different cultures, and the familiarity with norms, customs and values are among the benefits that migrants acquire abroad. As a subjective resource social capital can indeed be transferred, whereas as an objective resources it is locally embedded and therefore more difficult to transfer45.

Migrants can tap into social capital in both the sending and the receiving country. Benefits derived from such social capital include the information on jobs, social services, and housing abroad. But they also include information on these critical issues back home, and knowledge about business and investment opportunities, or loans to finance private enterprises. The more migrants have developed networking competencies, and have invested in maintaining social capital back home and in establishing social capital abroad, the more they have the freedom of choice to return, to stay away, or to shuttle back and forth46. But also, the more successful they should be in achieving their goals at both ends, because not only they have improved information, but also greater access to resources, and increased control and authority within the community. The return of this type of migrants has a greater likelihood to effect positive changes in the sending country.

Upon arrival in the receiving country, migrants establish social ties with persons or groups from their home country providing material support, help in making new social connections, and assistance to overcome their ignorance about the new community (Choldin, 1973). But then migrants gradually establish new webs of ties involving autochthonous and other foreign people

45 It has long been acknowledged that there is a territorial restriction of certain assets, a phenomenon that Da Vanzo (1981: 118) has denominated ‘location specific capital’. For Faist (1997: 204) the fact that social capital is difficult to transfer abroad raises the expectation that ‘potential migrants prefer those forms of movement that allow them to keep their social ties intact (circular migration), to interrupt them only briefly (seasonal migration) or to transfer the whole set of important social ties abroad (e.g. family migration in the context of chain migration’. 46 Faist (2000) notes that all three forms of capital can accrue the autonomy of individuals and enlarge their freedom of choice.

and start building corresponding social capital. In the sending country, social capital is maintained by migrants through visits, contacts by mail or telephone, remittances, marriage with a compatriot, or membership in associations linking receiving and sending countries. Presumably, both forms of social capital play a critical role because it is assumed that social capital is used to transfer and re-transfer back other forms of capital, e.g., financial and human capital (Faist, 1997). Interfacing both forms of social capital help migrants develop transnational identities and loyalties.

6. Elements for a meso-level approach to migration, return and development Past theoretical and empirical migration studies have generally taken either a macro-level or a micro-level analytical perspective. Macro theory and empirical research have focused on migration flows and chiefly on their volume, direction, composition and persistence. The implications of structural characteristics of the areas of origin and destination have been emphasised as well as their socio-economic and political attributes. This approach has paid particular attention to aspects that are common to social groups that share some economic, demographic, social or cultural attribute. Micro-level analyses have instead focused on the normative and psycho-social factors that determine the behaviours of individuals and influence their decision-making. From this perspective, personal and subjective aspects as for example perceptions, motivations, aspirations, expectations and values assume primary importance.

Both micro and macro-level should of course be carefully examined. Additional attention deserves, however, to be focused on the meso-level, which provides a crucial means to link these more conventional levels of analysis (Faist, 2000). At the meso-level, not only families and households, but also webs of kinship and other social clusters help to influence migrants’ decisions to move, stay or return. These decisions are rarely made by isolated individuals. Rather, economic, political and cultural institutions shape opportunities and constraints. Social capital, consisting of a set of specific resources that can be mobilised within groups, networks and organisations, can become an essential element for the analysis of migration trends.

21

In comparison with the other two approaches, the meso-level approach to migration has emerged more recently, in two strands of literature. Firstly, the processes of immigrant incorporation have been studied in economic sociology (Portes, 1995). Immigrants are not seen simply as isolated individuals, but rather as members of groups and participants in broader social structures that affect their mobility in various ways. Secondly, migrant networks have been studied where these are conceived as ‘sets of interpersonal ties that connect migrants, former migrants and non-migrants in origin and destination areas through ties of kinship, friendship, and shared community origin’ (Massey et al., 1993: 448)47.

A focus on meso-level processes stresses the connectivity of migrants and migration streams in the wider world of population groups and socio-economic, political and cultural processes. However, it does not obviate the need to draw boundaries around the specific migration phenomena of concern in any individual study, or to highlight particular comparisons or hypotheses that are relevant in specific situations. For example, in the research project on transnational migration, return and development in West Africa referred to in the first section, key boundaries and comparisons include:

• A focus on intercontinental migration from south to north, and return movements associated with this migration;

• A comparison between two cases studies, in Côte d’Ivoire and Ghana, in this case to examine the distinctiveness of processes that have emerged in contrasting colonial and post-colonial contexts;

47 Social networks based on family, friendship and community ties are important conduits of information and social and financial assistance for migrants (Massey et al., 1987). Such networks can explain immobility and migration, return migration and the perpetuation of migration because they help reducing the costs and risks of movement, and increasing the expected returns to migration. They are maintained by visits back home and return migration, associations and events linking sending and receiving countries, marriages which sustain social ties and obligations across time and space, and remittances (Boyd, 1989). Social networks are embedded in migration systems (Kritz and Zlotnik, 1992), and represent only one of several forms of linkage between receiving and sending countries (Fawcett, 1989).

• A distinction between skilled, elite migrants on the one hand, and poorer, unskilled migrants on the other, to draw out the extent to which the emergence of meso-level networks and institutions, as well as individual decision-making factors, varies between the two circumstances;

• A concentration on certain impacts of migration and return, including in this case the objective of poverty reduction, and the promotion of sustainable development.

Within such boundaries, it might be considered important to examine explicitly the short, medium and long term implications of migration and return. For example, in the short term, unskilled migrants might be expected to prioritise income generation, with the return of financial capital being a major impact on both poverty reducation and sustainable development. In turn, after a period of several years abroad, it might be expected that this ‘unskilled’ migrant would begin to learn new skills such that ‘human capital’ transfers become more important as a development impact, whilst in the longer-term, concern over the occupational prospects of offspring may consolidate processes of network building that ensure that ‘social capital’ becomes a more salient factor in chosing to return, or interact with the home country. The intention here is not to state unequivocally that such distinctions exist, or are the most salient variations between migrants of different vintages; but rather to illustrate the ways in which different hypotheses can emerge which draw together macro-, meso- and micro-level factors.

In the sub-sections that follow, a number of key working hypotheses of concern in the West African case study are reviewed.

6.1 Survival and mobility strategies

Starting with the micro-level, a number of questions emerge from migrants’ own experience which are of relevance to explaining the extent of return, and the relevance of migration and return to development (including poverty reduction and sustainable development) in the country of origin. These can be roughly categorised into experiences whilst abroad, and experiences on return:

22

The migrants’ experience abroad.

Whilst abroad, there are a number of features of the individual experiences of migrants and their families which will influence both their propensity to return, and the extent of any ongoing involvement they maintain with their country of origin. It is worth exploring the extent to which migrants of different types (skilled, unskilled, from certain countries, short vs. long term etc.) are more or less likely to return, and/or acquire the ability or desire to contribute towards poverty reduction and development initiatives in their home communities whilst overseas. It is also important to examine where the origin of such tendencies lie. Are particular groups of migrants more likely to develop innovative strategies to gain income, knowledge or valuable social networks, and what kinds of institutional and legislative contexts frame or enhance such strategies?

For example, given that there is known to be gender selectivity in the migration process, does that lead women and men to have different migratory experiences, with consequently different impacts on return and development? Given that unskilled migrants are more likely to have worked in unskilled jobs abroad, they may be less likely to bring back skills that help to boost local economies. In contrast, skilled migrants are perhaps more likely to acquire additional human, financial and social capital whilst they are abroad. Also, and stressing the inter-relationship of the micro, meso and macrolevels, it is worthwhile to consider evidence of migrants or their associations engaging with initiatives to promote micro-enterprises, employment, or community development projects; some notable examples are ‘home town’ associations in the Nigerian context, and the work of Muslim brotherhoods in Senegal (Riccio, 2000) – are such initiatives related to specific kinds of migrant experience, or specific sending contexts?

As was noted in section 4.4, Rogers (1984) developed a series of hypotheses about what influences propensity for migrants to return (Box 1). Although these factors reflect processes operating at macro, meso and micro level, they are conceptualised from the point of view of the micro decision-making process for individual migrants and their families, and in this sense are amenable to examination through individual interviews, questionnaires and discussions.

For Rogers (1984), there is not a clear distinction between events taking place at the aggregate and individual levels. Macro-level influences work themselves out through the perceptions and decisions of migrants acting at the micro-level. Migrants have varying degrees of awareness of changes occurring at the structural and systemic level. They also have quite different sets of motives for return which are influenced by differences in socio-economic and demographic characteristics such as age, sex, education, occupation, and socio-economic status. This acknowledgement makes Rogers (1984: 291) argue that ‘observations about return populations being ‘positively’ or ‘negatively’ selected from among migrant populations or citizens of a home country really gives us only rather little

Box 1: Reasons for return migration (after Rogers 1984: 288-90)

1. Perception that positive changes have occurred in those situations in their home countries that brought about or contributed to the original migration.

2. Events in home countries unrelated to the original migratory intentions necessitate, in the migrant’s view, their presence in the country.

3. Host country no longer satisfies the needs that they came to satisfy, or satisfies them less well than earlier.

4. Events in the host countries unrelated to the motives for migration make it less desirable for them to remain there than it had been before.

5. Satisfaction of the goals that the migrant or their families had set or, if they had no specific targets, because they feel that they can now change their original situation in the home country.

6. Families at home need the migrant, and they agree to follow their call.

7. Migrants consider they erred in their assessment that the needs that motivated their migration could be satisfied better in the host country than at home.

8. Although the specific needs that motivated the migration are fulfilled in the home country, they feel that the costs involved are too high.

23

information and can even be misleading in their implication’. One of her suggestions is to analyse the full range of cases in order to ascertain specific patterns of return, and the frequency with which they occur.

The migrants’ experience on return.

In addition to different experiences whilst abroad, different groups of migrants are likely to have varying experiences on their return – and these too may play a major role in influencing the nature of the impact of migration on poverty reduction and sustainable development. For example, how do migrants adapt and readjust back home? What obstacles do they encounter in utilising their newly acquired capital in their home economies and societies? What resistance do they face in the workplace, family, and social environments?

Examination of both the opportunities and obstacles faced by returning migrants can help us to understand their role both directly in combating poverty or promoting sustainable development, and indirectly in the design and implementation of policies or initiatives aimed at these objectives. Return migrants may make a crucial contribution to the development of micro and small-scale businesses, and/or employment creation for poor people, whilst they may also help to promote good governance and management excellence in the public and private spheres.

Yet once again, it is salient to ask how the roles of skilled and unskilled returnees, or people returning to different legislative or national contexts differ. Not only have skilled migrants had different experiences abroad, but they are also likely to be perceived differently from unskilled migrants on their return, with this difference also varying according to whether the place of return is a rural or urban areas. Unskilled migrants may have had a direct impact on poverty alleviation through easing labour market imbalances and promoting remittances, and their definitive return to their place of origin may spell an end to these local benefits. In contrast, skilled migrants may be able to go back to positions of responsibility and authority in the public and private sectors of their home country, and thus have more influence over development trajectories.

Overall, it is worth enquiring whether there are certain optimal lengths of absence for migrants to

have a maximum benefit on issues such as poverty reduction and sustainable development, and whether this varies for different groups of migrants. For example, if the length of time spent away is too long, returnees may no longer know their home country and make important miscalculations in the way in which they invest various forms of financial, human and social capital acquired. On the other hand, if their time away is too short, there may simply not have been the time for migrants to acquire the different forms of capital that could impact on development.

6.2 Structural and institutional constraints

The previous section focused on the mobility and survival strategies of immigrants and how different forms of experience of different migrant groups may impact on propensity to return, and the impact of this return. However, it is difficult to isolate these factors from structural and institutional constraints (and opportunities) which also inflence the nature of migrant impact on development. For example, Bovenkerk’s list of such factors, referred to in section 5, includes elements that could be considered both ‘micro’ and ‘macro’ (Box 2), although on the whole they point us towards structural explanations of the varied impact of different migrant flows.

Box 2: Factors affecting potential of migration and return to favour development (after Bovenkerk 1974: 46-9)

1. Number of returnees, in absolute and relative terms;

2. Concentration of returnees in time, because when migrants return are concentrated in a shorter time span, the critical mass exists that is needed to bring about change;

3. Duration of absence, because when it is too short, not enough may be learned to transfer anything meaningful, and when it is too long, migrants' may become too detached from home or too old to translate new ideas into practice;

4. Social class of the migrants, as skilled elite migrants seem more likely to assume the role of agents of change than unskilled labour migrants;

5. Motives for return, because the more return migrants have responded to pull-factors in their

24

home countries the greater the chance for innovation, the more they have reacted to push-factors the less the chance for innovation;

6. Degree of difference between the country of emigration and the country of immigration, because if this is too great the skills and experiences that migrants acquire abroad may not be useful back home;

7. Nature of the acquired training and skills, as it seems that the more general the training received abroad the greater chances for innovation, the more specific the training the less transferable in the home setting;

8. Organisation of return, because the better the return is planned and organised the greater the chance for change;

9. Political relationship between the countries of emigration and return, as the definition of progress in the home country may influence returnees' impacts.

It is worth noting that the nature, strength and form of migration differs according to the situation in both the sending countries, and the receiving countries. For example, in the case of highly skilled migration, because of adverse economic and political conditions in Ghana since the 1960s, Ghanaians have tended to emigrate more frequently and for longer periods than skilled personnel in neighbouring Côte d’Ivoire. In turn, better work opportunities in Côte d’Ivoire have led many Ivorians to choose only to study abroad, whereas Ghanaians study and work abroad. Moreover, the more dynamic private sector in Côte d’Ivoire has led many returnees to go back to existing firms, whereas Ghanaians have arguably been forced to set up their own businesses if they are to invest the capital that they have acquired. Similar comparisons could be explored for the case of unskilled migrants and returnees.

Of course, if the economic, social, cultural political differences between the sending and the receiving country are not great, it could also be argued that there will be relatively little potential or incentive for the promotion of change on the part of migrants,whilst if the difference is too great there will be much resistance to efforts to change the host society on behalf of the return migrant, on the grounds that the task is seen as

too large. However, structural and institutional constraints may work at a more direct level than this: as well as wider differences in the local economy, it is worth asking whether government policies specifically orientated towards migration and return can influence both the pattern of return, and/or the impact of migration and return on development processes.

6.3 Migration, return and social networks

Moving beyond issues of individual experience on the one hand, and structural and institutional constraints on the other, there remain a number of important questions that relate to the influence of meso-scale institutions. Of key interest here is the way in which markets and other institutions, which mediate the different forms of capital mentioned above, operate to promote certain consequences of migration and return through capital transfer.

For example, the development of social networks and ties can be seen as instrumental not only for emigration, but also for return (to find satisfactory employment, suitable housing, etc.) or the development of transnational activities (contribution to home town associations, maintenance of cultural links between home and receiving countries, etc.). One possible hypothesis is that the stronger the social and symbolic ties of migrants to persons and/or institutions in the country of origin, the higher the propensity to return, and the greater the chances for this return to effect social and economic change. Meanwhile, it is often assumed that the stronger the social and symbolic ties of migrants to persons or institutions in the emigration country, the lower the propensity to return. However, when those with strong ties to the emigration country do return, or for that matter develop transnational activities that support home areas without permanently returning, it may be that they are in an even stronger position to effect social change (Al-Ali et al., 2001). This can be seen as related to migrants’ ability to either manipulate, or learn from the use of strong social networks whilst abroad.

There has been a fair focus on the literature on capital transfers to date which emphasizes the role of financial (and to a lesser extent human) capital, although even in this case, an interesting question still remains for particular case studies as to whether this capital is actually invested in activities that address poverty for specific sections

25

of the home population. However, it is in the area of social capital transfers, via social networks, that there is least current information; whilst at the same time, a broader question remains as to whether permanent return is a prerequisite for transfers of human, social and financial capital to occur.

7. Linking micro and macro perspectives Such questions need to be answered at different levels of analysis. In this sense, it is important to draw on a cross-sectional and a longitudinal perspective to allow comparison between countries and different migrant groups. This should help to examine migration processes and their local developmental implications and to explore the extent of involvement of migrants and returnees in poverty reduction and sustainable development initiatives.

7.1 Methods

Various methodologies are available to gain a longitudinal perspective. In particular, qualitative methodologies, include narratives to reconstruct family and personal life histories, interviews using open-ended questions regarding individual experiences and expectations for the future, and direct observation to deepen and enrich the understanding of the role of migration and return in each contextual setting. Analysis of information from LSMS data sets can also be used to derive time series information on migration and remittances. Meanwhile, from a cross-sectional perspective, quantitative methods are of wider use to obtain empirical data concerning the socio-demographic characteristics of particular populations of returned migrants and migrants who remain abroad.

In conducting a questionnaire survey of migrants, or return migrants, a number of pitfalls arise. Most obviously, a sample frame is lacking. One response to this has been the 'ethnosurvey' framework of Massey and Zenteno (1999: 766), which involves random sampling of households in known sending communities, carried out at a time when seasonal migrants are known to be back in Mexico, and supplemented by non-random samples of migrants in US destination areas contacted through connections established in

Box 3: Fieldwork plans in the ‘Transrede’ project

In planned research in Ghana and Côte d’Ivoire, fieldwork will be carried out in three phases. During a pilot phase, after a review of documentary evidence, semi-structured, partially standardised and largely pre-coded questionnaires will be finalised, and pre-tested with a limited number of highly skilled and unskilled returnees. With the elite group in particular, it is felt that closed questions are appropriate, as this group are likely to have limited time. This makes testing of the questionnaires, and training of local researchers in partner organisations of particular importance.

In the second phase, the survey will then be administered face-to-face to a sample of about 100 – 150 highly skilled and a similar number of unskilled returnees in Ghana and Côte d’Ivoire. Interviewees will be selected through contact with consular offices, institutions of higher education, migrant associations, and other bodies where return migrants might be expected to concentrate. A rolling (or ‘snowball’) sampling procedure will be adopted, in which returnees identified through as diverse a range of mechanisms as possible will then be asked to provide names and contact addresses of other returned migrants that they know. The limitations of this technique will be borne in mind, and responded to through careful choice of initial and subsequent contacts.

In the third phase, a series of in-depth interviews and focus groups, some tape-recorded, will be conducted with returnees in each country of origin, as well as with potential returnees in London, Paris, and Washington, DC-Baltimore area, all locations in which there are known to be significant numbers of Ghanaian and Ivorian migrants. These are seen as providing an important opportunity to discuss the results of the survey of returned migrants, as well as gathering new information on attitudes of returnees and those who remain abroad towards the relationship between migration and poverty reduction. Other key informants, such as CEOs of private companies, high-level officials in government and international organisations, heads of NGOs, religious and political leaders, and others will also be interviewed to explore particular policy implications of the study.

26

Mexico. Such an approach could also pick up returnees, although it might be more appropriate for unskilled migrants coming from and returning to rural areas, than for skilled or intercontinental migrants originating in capital cities.

An alternative strategy of starting with snowball samples of respondents in receiving areas, and then following leads to the country origin, has proved fruitful in recent work at Sussex and University College London on Bosnian and Eritrean refugees (Al-Ali et al., 2001). However, this study was orientated primarily to migrants who did not return (at least not permanently), rather than those that did. Meanwhile, survey work currently being undertaken in Asia by the Asia-Pacific Migration Network provides a model for questionnaire design – with the use of similar questions potentially facilitating the cross-referencing of results. This survey explores the actual practice of migration; the extent of involvement in transfers of various kinds of capital either whilst away or on return; and the uses to which this capital is put. All of these are relevant to the linkages between migration, return and development in West Africa.

7.2 Policy

It is also seen as essential to link on-going research on migration, return and development to emerging policy initiatives in the field. As has been alluded to above, existing policy initiatives in this field have been orientated primarily towards encouraging return migration – especially of the highly skilled (Ghosh, 2000b). This has been seen as the most effective way of reversing the brain drain, and mobilising the resources of migrants for development back home. Yet the argument presented in this paper suggests there is much uncertainty about the impacts of migration and return on development. At the same time, policies to support the return of migrants have often been seen as disappointing at best (Koser, 2001; Black, 2001), or worse, motivated primarily by exclusion from the north, rather than a commitment to development in the south. In this context, further research in this field, including research focused on the return process, should be able to identify alternative routes for policy, that build on different conceptualisations of the relationship between migration, return and development.

A first point is that this paper has highlighted how differences in context can affect outcomes from

the return process in terms of development. Typologies of return, referred to in section 4.3, might be expanded to emphasize that it is not only the characteristics of returnees that are important to the return process (such as the skills possessed), but also the context of host and return countries research. Even with considerable motivation, a returnee is unlikely to make a successful contribution to his or her home country unless some form of capital has been generated overseas that can be put to use. Meanwhile, even with capital, conditions need to exist in the country of return that allow efficient investment of that capital.

The key issue here that is of importance to policy-makers concerns identification of contexts that favour investment of resources on return, as well as the types of capital that can be best invested in development in different circumstances. For example, whilst attention has clearly been paid in return schemes to the transfer of human capital, there has been relatively little investigation of the relative merits of repatriation of financial capital through remittances, and the various ways in which this capital is transferred (or not) on return. Meanwhile, the study of social capital in migration remains in its infancy, and there are few pointers at present as to how public policy can link into migrants’ and returnees’ social capital and make these work for the development process.

Another important point concerns the way in which public policy can adjust to new patterns of migration in an increasingly ‘transnational’ age. If, has been suggested, return migration no longer represents the ‘closure’ of the migration cycle, but a stage along a process of increasingly fluid movements between countries, what does this say about the type of public policy that is required to mobilise the resources of such migrants? A traditional viewpoint is to see rootedness in a particular place as representing commitment to that place – hence when people migrate, they are seen as a ‘loss’ to the sending country, with only their permanent return reversing that loss. However, emerging research on transnationalism highlights how people may make major contributions, economically, socially, politically, or culturally, to their place of origin, without returning permanently (Glick-Schiller et al., 1992; 1995). How to mobilize the potential of such transnational migrants, and what kind of return(s) can be important in stimulating the process of re-engagement with a country of

27

origin, represent important areas for emerging research.

As migrants become increasingly flexible and fluid in their movements and actions, there is growing recognition that both research and public policy need to adapt to cope with such changes. That there have been public policy shifts is undoubted – as countries such as the Philippines (Jones and Findlay, 1998) and Eritrea (Al-Ali et al., 2001) move to mobilize ‘their’ workforce overseas, and bring them into the development process. But how these processes play out for skilled and unskilled migrants; the extent to which return to country of origin – whether permanent or temporary – can or should play a role in such processes; and the attitude that should be taken by receiving countries and non-governmental organisations concerned with development, all remain issues in urgent need of further work.

Bibliography Abadan-Unat, N. (1976) 'Migration as an obstacle for re-integration in industry: The Turkish case', Studi Emigrazione, 13(43), 319-34.

Abraham, P.M. (1968) 'Regaining high level Indian manpower from abroad', Manpower Journal, 3(4), 83-117.

Adepoju, A. (1998) ‘Linking population policies to international migration in sub-Saharan Africa’, in R. Appleyard (Ed) Emigration Dynamics in Developing Countries, Volume I: Sub-Saharan Africa, Aldershot: Ashgate, 301-37.

Adepoju, A. (1995a) ‘Emigration dynamics in Sub-Saharan Africa’, International Migration, 33(3/4): 315-91.

Adepoju, A. (1995b) ‘Migration in Africa: an overview’, in J. Baker and T.A. Aina (Eds) The Migration Experience in Africa, Sweden: Nordiska Afrikainstitutet, 87-108.

Adepoju A. (1994) ‘Preliminary analysis of emigration dynamics in sub-Saharan Africa’, International Migration 32(2): 197-216.

Adepoju, A. (1991a) ‘South-North migration: the African experience’, International Migration, 29(2): 205-223.

Adepoju, A. (1991b) ‘Binational communities and labor circulation in sub-Saharan Africa’, in D.G. Papademetriou and P.L. Martin (Eds) The Unsettled Relationship: Labor Migration and Economic Development, Westport: Greewood Press, 45-64.

Adepoju, A. (1988a) ‘Links between internal and international migration: the African situation’, in C. Stahl (Ed) International Migration Today, Volume 2: Emerging Issues, Paris: UNESCO, 34-58.

Adepoju, A. (1988b) ‘International migration in Africa south of the Sahara’, in R. Appleyard (Ed) International Migration Today, Volume I: Trends and Prospects, Paris: UNESCO, 17-88.

Ahmed, I. (2000) ‘Remittances and their economic impact in post-war Somaliland’, Disasters, 24(4), 380-89.

Al-Ali, N., Black, R. and Koser, K. (2001) ‘The limits to ‘transnationalism’: Bosnian and Eritrean refugees in Europe as emerging transnational communities, Ethnic and Racial Studies, 24(4)

Al-Rasheed, M. (1992) ‘Political migration and downward socio-economic mobility: the Iraqi community in London’, New Community, 18(4): 537-49.

Ammassari, S. (1994) 'Internal migration, population distribution and development policy', IDS Discussion Paper, 342, Brighton: Institute of Development Studies.

Anwar, M. (1979) The Myth of Return: Pakistanis in Britain, London: Heinemann.

Appleyard R. (Ed)( 1998) Emigration Dynamics in Developing Countries, Volume I: Sub-Saharan Africa, Aldershot: Ashgate.

Appleyard, R. (1992) ‘International migration and development – an unresolved relationship’, International Migration, 30(3/4), 251-66.

Appleyard, R. (1989a) ‘Migration and development: myths and reality’, International Migration Review, 23(3): 486-505.

Appleyard, R. (1989b) (Ed) The Impact of International Migration on Developing Countries, Paris: OECD.

28

Appleyard R. (Ed) (1988) International Migration Today, Volume I: Trends and Prospects, Paris: UNESCO.

Ardittis, S. (1991) ‘Targeted reintegration of expatriate brains into developing countries of origin: the ECC-IOM experience in Central America’, International Migration, 33(1): 371-88.

Ardittis, S. (1985) ‘The assisted return of qualified migrants to their countries of origin: The UNDP and ICM multilateral programmes’, International Migration for Employment, Working Paper, Geneva: ILO.

Arif, G.M. and Irfan, M. (1997) ‘Return migration and occupational change: the case of Pakistani migrants returned from the Middle East’, Pakistani Development Review, 36(1): 1-37.

Arthur, J.A. (1991) ‘International labour migration patterns in West Africa’, African Studies Review, 34(3): 65-87.

Bhagwati, J.N. (Ed) (1976) The Brain Drain and Taxation: Theory and Empirical Evidence, ll, Amsterdam: North-Holland Publishing company.

Baker J. and Aina T.A. (Eds) (1995) The Migration Experience in Africa, Sweden: Nordiska Afrikainstitutet.

Balán, J. (1988) ‘Selectivity of migration in international and internal flows’, in C. Stahl (Ed) International Migration Today, Volume 2: Emerging Issues, Paris: UNESCO, 46-62.

Bennell, P.S. and Godfrey, M. (1983) ‘The professions in Africa: some interactions between local and international markets’, Development and Change, 14(3), 372-402.

Bernard, H.R. and Comitas, L. (1978) ‘Greek return migration’, Current Anthropology, 19(3), 658-59.

Bilsborrow, R.E., Oberai, A.S., and Standing, G. (Eds)(1984) Migration Surveys in Low Income Countries: Guidelines for Survey and Questionnaire Design, London: Croom Helm.

Black, R. (2001) ‘Return and reconstruction: missing link or mistaken priority in post-Dayton Bosnia and Herzegovina?’ SAIS Review, forthcoming

Boyd, M. (1989) ‘Family and personal networks in international migration: recent developments and new agendas’, International Migration Review, 23(3): 638-70.

Boyle P., Halfacree, K. and Robinson, V. (Eds) (1998) Exploring Contemporary Migration, New York: Longman.

Böhning, W.R. (1986) Bibliography on International Return Migration, International Migration Employment Working Paper 16, Geneva: International Labour Office.

Böhning, W.R. (1984) Studies in International Labour Migration, London : Macmillan.

Böhning, W.R. (1975) 'Some thoughts of emigration form the Mediterranean basin', International Labour Review, 111(3), 251-77.

Böhning, W.R. (1972) ‘The social and occupational apprenticeship of Mediterranean workers in West Germany’, in The Demographic and Social Pattern of Emigration from the Southern European Countries, M. Livi-Bacci (Ed.), Florence: Università di Firenze, 175-259.

Bourdieu, P. and Wacquant, L. (1992) An Invitation to Reflexive Sociology, Chicago: University of Chicago Press.

Bovenkerk, F. (1982) ‘Why returnees generally do not turn out to be ‘agents of change’’, in Return Migration and Remittances: Developing a Caribbean Perspective, W.F. Stinner et al., RIIES Occasional Papers 3, Washington D.C.: Smithonian Institution.

Bovenkerk, F. (1974) The Sociology of Return Migration: A Bibliographic Essay, Publications of the Research Group on European Migration Problems 20, The Hague: Martinus Nijhoff.

Brydon, L. and Gould, W.T.S. (1984) ‘International migration of skilled labour within West Africa: A review and agenda for future work’, Liverpool Papers in Human Geography, Working Paper 17.

Caldwell, J.C. (1969) African Rural-Urban Migration: The Movement to Ghana’s Towns, Canberra: Australian National University Press.

Callea, S. (1986) 'Different forms, reasons and motivations for return migration of persons who

29

voluntarily decide to return to their countries of origin', International Migration, 23(1), 61-71.

Castles, S. and Miller M.J. (1998) The Age of Migration: International Population Movements in the Modern World, Second Edition, London: Macmillan.

Castles, S. and Kosack, G. (1973) Immigrant Workers and Class Structure in Western Europe, London: Oxford University Press.

Chepulis, R. (1984) ‘Return migration: An analytical framework’, in D. Kubat (Ed) The politics of Return: International Return Migration in Europe, New York: Center for Migration Studies, 239-45.

Chilivumbu, A. (1985) Migration and Uneven Rural Development in Africa: The Case of Zambia, Lanham: University Press of America.

Choldin, H.M. (1973) 'Kinship networks in the migration process', International Migration Review, 7(2), 163-75.

Chukunta, N.K.O. (1979) ‘Introduction’, Issue, 9(4), 3-5.

Condé, J. (1989) ‘Measures to encourage return migration and reintegration of returned migrants in their home country’, in R. Appleyard (Ed) The Impact of International Migration on Developing Countries, Paris: OECD, 323-32.

Connell, J., Dasgupta, B., Laishley, R. and Lipton M. (1976) Migration from Rural Areas, Dehli: Oxford University Press.Cordell, D.D., Gregory, J.W. and Piché, V. (Eds) (1996) Hoe and Wage: A Social History of a Circular Migration System in West Africa, Boulder: Westview Press.

Dahya, E. (1973) 'Pakistanis in Britain, transients or settlers?', Race, 14(3), 241-77.

Das, M.S. (1979) ‘Brain drain controversy and African scholars’, Issue, 8(4), 27-30.

Da Vanzo, J. (1981) ‘Microeconomic approaches to studying migration decisions’, in G.F. De Jong and R.W. Gardner (Eds), Migration Decision Making, 90-129.

Diatta, M.A. and Mbow, N. (1999) ‘Releasing the Development Potential of Return Migration: the

case of Senegal’, International Migration, 37(1): 243-66.

D’Oliveira e Sousa, J. (1989) ‘The brain drain issue in international negotiations’, in R. Appleyard (Ed) The Impact of International Migration on Developing Countries, Paris: OECD, 197-212.

Dumon, W. (1984) 'Problems faced by migrants and their family members, particularly second generation migrants, in returning and reintegrating into their countries of origin', in D. Kubat (Ed) The politics of Return: International Return Migration in Europe, New York: Center for Migration Studies, 95-128.

Esser, H. (1988) Aspekte der Wanderungssoziologie: Assimilation und Integration von Wanderern, Ethnischen Gruppen und Minderheiten. Eine Handlungstheoretische Analyse, Darmstadt: Luchterhand.

Faist, T. (2000) Volume and Dynamics of International Migration, Oxford: Oxford University Press.

Faist, T. (1997) ‘The crucial meso-level’, in T. Hammar, G. Bochmann, K. Tamas and T. Faist (Eds) International Migration, Immobility and Development. Multidisciplinary Perspectives, Oxford: Berg, 187-217.

Fawcett, J.T. (1989) ‘Networks, linkages and migration systems’, International Migration Review, 23(3): 671-80.

Fei, J.C.H. and Ranis, G. (1961) ‘Theory of economic development’, The American Economic Review, 5(4).

Findlay, A. (1990) ‘ A migration channels approach to the study of high level manpower movements: A theoretical perspective’, International Migration, 28(1): 15-22.

Findley, S.E. (1995) ‘Migration and family interactions in Africa’, in A. Adepoju (Ed.) The African Family in the Development Process.

Findley, S.E. and Sow, S. (1998) ‘From season to season: agriculture, poverty and migration in the Senegal River Valley’, in R. Appleyard (Ed) Emigration Dynamics in Developing Countries, Volume I: Sub-Saharan Africa, Aldershot: Ashgate, 69-144.

30

Findley, S.E. and Williams, L. (1990) Women who go and women who stay: Reflections of family migration processes in a changing world, Working Paper Series, Geneva: ILO.

Fischer, P.A., Martin, R. and Straubhaar, T. (1997) ‘Interdependencies between Development and Migration’, in International Migration, Immobility and Development. Multidisciplinary Perspectives. Oxford: Berg, 91-132.

Foote, K.A., Hill, K.H. and Martin, L.G. (Eds) (1993) Demographic Change in Sub-Saharan Africa, Washington D.C.: National Academy Press.

Gardiner, R.K.A. (1968) ‘Africa’, in W. Adams (Ed) The Brain Drain, New York: Macmillan, 194-202.

Ghosh, B. (1992) ‘Migration-development linkages: some specific issues and practical policy measures’, International Migration, 30(3/4), 423-452.

Ghosh, B. (2000a) ‘Introduction’, in B. Ghosh (Ed) Return Migration. Journey of Hope or Despair? Geneva: IOM/UN.

Ghosh, B. (2000b) ‘Return Migration: reshaping policy approaches’, in B. Ghosh (Ed) Return Migration. Journey of Hope or Despair? Geneva: IOM/UN.

Gilani, I.S. (1988) ‘The effects of emigration and return on sending countries: the case of Pakistan’, in C. Stahl (Ed) International Migration Today, Volume 2: Emerging Issues, Paris: UNESCO, 204-16.

Glaser, W. (1978) The Brain Drain. Emigration and Return, UNITAR Research Report 22, London: Pergamon Press.

Glick-Schiller, N., Basch L. and Szanton-Blanc C. (Eds) (1992) Toward a transnational perspective on migration, New York: New York Academy of Science.

Glick-Schiller, N., Basch, L. and Szanton-Blanc, C. (1995) 'From immigration to transnational migration: theorizing transnational migration', Anthropological Quarterly, 68(1), 48-63.

Gmelch, G. (1992) Double passage: the Lives of Caribbean Migrants Abroad and Back Home, Ann Arbor: University of Michigan Press.

Gmelch, G. (1987) ‘Work, innovation and investment: the impact of return migrants in Barbados’, Human Organization, 46(2): 131-40.

Gmelch, G. (1980) ‘Return Migration’, Annual Review of Anthropology, 9: 135-59.

Gmelch, G. (1979) 'Irish return migration: the socio-demographic characteristics of return migrants', Papers in Anthropology, 20(1), 155-66.

Gould, W.T.S. (1988a) ‘Government policies and international migration of skilled workers in sub-Saharan Africa’, Geoforum, 19(4), 433-45.

Gould, W.T.S. (1988b) ‘Skilled international migration: An introduction’, Geoforum, 19: 381-85.

Gould, W.T.S. (1974) ‘International migration of skilled labour within Africa: A bibliographical review’, International Migration Review, 8(3): 347-65.

Grey-Johnson, C. (1986) ‘Measures to facilitate the return and reintegration of highly skilled migrants into African countries’, International Migration, 23(1): 197-211.

Hammar, T., Brochmann, G., Tamas, C. and Faist, T. (Eds) (1997) International Migration, Immobility and Development. Multidisciplinary Perspectives, Oxford: Berg.

Harris, J.R. and Todaro M.P. (1970) ‘Migration, unemployment and Development: a two-sector analysis’, American Economic Review, 60, 126-42.

Hermele, K. (1997) ‘The discourse on Migration and Development’, in Hammar et al. (Eds) International Migration, Immobility and Development. Multidisciplinary Perspectives, Oxford: Berg.

Hernandez-Alvarez, J. (1968) 'Migration, return and development in Puerto Rico', Economic Development and Cultural Change, 16(4), 574-87.

Hernandez-Alvarez, J. (1967) Return Migration to Puerto Rico, Berkley: University of California.

Hoffmann-Nowotny, H.-J. (1970) Migration - ein Beitrag zu einer soziologischen Erklärung, Stuttgart: Enke.

31

Hoffmann-Nowotny, H.-J. (1973) Soziologie des Fremdenarbeiterproblems: Eine theoretische und empirische Analyse am Beispiel Schweiz. Stuttgard: Ferdinand Enke.

International Organization for Migration (1991) Ninth IOM Seminar on Migration: South-North Migration, Geneva: IOM.

Jones H. and Findlay, A. (1998) ‘Regional economic integration and the emergence of the East Asian international migration system’, Geoforum, 29(1): 87-104.

Keely, C.B. (1986) ‘Return of talent programs: rationale and evaluation criteria for programs to ameliorate a ‘Brain Drain’’, International Migration, 23(1), 179-95.

King, R. (2000) ‘Generalizations from the history of return migration’, in B. Ghosh (Ed) Return Migration. Journey of Hope or Despair? Geneva: IOM/UN.

King, R. (1996) ‘Migration in a worlds historical perspective’, in J. van der Broeck (Ed) The Economics of Labour Migration, Cheltenham: Edward Elgar, 7-76.

King, R. (1986) Return Migration and Regional Economic Problems, London: Croom Helm.

King, R. (1986) ‘Return migration and regional economic development: An overview’, in R. King (Ed) Return migration and Regional Economic Problems, London: Croom Helm.

King, R. et al. (1983) ‘Return Migration: A Review of the Literature’, Discussion Papers in Geography, 19, Oxford: Oxford Polytechnic.

King, R. (1978) ‘Return migration: a neglected aspect of population geography’, Area, 10(3): 175-83.

King, R., Strachan A. and Mortimer J. (1986) ‘Gastarbeiter go home: return migration and economic change in the Italian mezzogiorno’, in R. King (Ed) Return Migration and Regional Economic Problems, London: Croom Helm, 38-68.

Koser, K. (2001) ‘The return and reintegration of rejected asylum-seekers and irregular migrants: an analysis of government-assisted return programmes in selected European countries’, IOM Migration Research Series, no. 4, Geneva: IOM

Koser, K. (2000) ‘Return, readmission and reintegration: changing agendas, policy frameworks and operational programmes’, in B. Ghosh (Ed) Return Migration. Journey of Hope or Despair? Geneva: IOM/UN.

Krane, R.E. (1976) 'Effects of international migration upon occupational mobility, acculturation and the labour market in Turkey', in Manpower Mobility Across Cultural Boundaries: Social, Economic and Legal Aspects. The case of Turkey and West Germany. Social, Economic and Political Studies of the Middle East, 16, 161-204.

Krane, R.E. (1973) ‘The effects of cyclical international migration upon socio-economic mobility’, International Migration Review, 7(4): 427-36.

Kritz, M.M., Lim, L.L. and Zlotnik H. (Eds) (1992) International Migration Systems. A Global Approach, Oxford: Claredon Press.

Kritz, M.M. and Zlotnik H. (Eds) (1992), ‘Global interactions: migration systems, processes, and policies’, in M.M. Kritz et al. (Eds) International Migration Systems. A Global Approach, Oxford: Claredon Press.

Lee, E.S. (1966) ‘A Theory of Migration’, Demography, 3(1): 47-57.Lepore, S. (1986) 'Problems confronting migrants and members of their families when they return to their countries of origin', International Migration, 23(1).

Lewis, A.W. (1954) ‘Economic development with unlimited supplies of labour’, The Manchester School of Economic and Social Studies, 22(2).

Lewis, G.J. (1982) Human Migration. A Geographical Perspective. London: Croom Helm.

Libercier, M.-H. and Schneider, H. (1996) Migrants. Partners in Development Co-operation, Paris:OECD.

Lipton, M. (1982) Migration from rural areas in poor countries: the impact on rural productivity and income distribution, World Development, 8(1).

Lipton, M. (1980) ‘Migration from rural areas of poor countries: the impact on rural productivity and income distribution’, World Development, 8(1): 1-24.

32

Livi-Bacci, M. (1993) ‘South-North migration: a comparative approach to North American and European experiences’, in The Changing Course of International Migration, Paris: OECD, 19-46.

Logan, B.I. (1992) ‘The brain drain of professionals, technical and kindred workers from developing countries: some lessons from Africa-US flow of professionals (1980-1989)’, International Migration, 30(3/4): 289-312.

Logan, B.I. (1990) ‘An assessment of the potential application of the Transfer of Knowledge through Expatriate Nationals (TOKTEN) Programme in sub-Saharan Africa, Applied Geography, 10, 223-26.

Mabogunje, A.L. (1970) ‘Systems approach to a theory of rural-urban migration’, Geographical Analysis, 2, 1-17.

Makinwa-Adebusoye, P.K. (1995) ‘Emigration dynamics in West Africa’, International Migration, 23(3/4): 435-68.

Massey, D.S., Arango, J., Hugo, G., Kouaouci, A., Pellegrino, A. and Taylor, E. (1993) ‘Theories of international migration: a review and appraisal’, Population and Development Review, 19(3): 431-66.

Massey, D.S., Alacron, R., Durond, J. and Gonzalez, H. (1987) Return to Atzlan: The social process of international migration from Western Mexico, Berkley: University of California Press.

Massey, D.S. and R. Zenteno (1999) ‘A validation of the ethnosurvey: the case of Mexico-U.S. migration’. International Migration Review 34(3): 766-93

Mundende, D.C. (1989) ‘The brain drain and developing countries’, in R. Appleyard (Ed) The Impact of International Migration on Developing Countries, Paris: OECD, 183-95.

Murillo Castaño, G. (1988) 'Effects of emigration and return on sending countries: the case of Columbia', in C. Stahl (Ed) International Migration Today, Volume 2, Emerging Issues, Paris: Unesco.

Muschkin, C.G. (1993) ‘Consequences of return migrant status for employment in Puerto Rico’, International Migration Review, 27(1): 79-102.

Nair, G. (1999) ‘Return of overseas contract workers and their rehabilitation and development in Kerala (India). A critical account of policies, performance and prospects’, International Migration, 37(1), 209-242.

Ndioro Ndiaye (2000) ‘Le phénomène de la fuite des cerveaux et le programme de retour de personnel qualifié mis en oeuvre par l’OIM en Afrique’, Paper presented to IOM – UNECA Brain Drain Meeting, Addis Ababa, 22-24 February 2000.

Obia, G.C. (1993) ‘Brain drain and African development: a descriptive framework for deriving indirect benefits’, Journal of Third World Studies, 10(2), 74-97.

OECD (2000) Globalisation, Migration and Development, Paris: OECD.

Oucho, J.O. (1995) ‘International migration and sustainable human development in Eastern and Southern Africa’, International Migration, 23(1): 31-51.

Oucho, J.O. (1990) ‘Migrant linkages in Africa: Retrospect and prospect’, in Conference on the Role of Migration in African Population Studies for the 90s, Commissioned Papers, Dakar: Union for African Population Studies (UAPS), 109-41.

Palmer, I. (1985) The Impact of Male Out-Migration on Women in Farming, West Hartford: Kumarian Press.

Papademetriou, D.G. and Martin, P.L. (1991) (Eds) The Unsettled Relationship: Labor Migration and Economic Development, Westport: Greewood Press.

Papademetriou, D.G. (1991) ‘Migration and development: the unsettled relationship’, in D.G. Papademetriouand P.L. Martin (Eds) The Unsettled Relationship: Labor Migration and Economic Development, Westport: Greewood Press.

Piore, M.J. (1979) Birds of Passage: Migrant Labor and Industrial Societies, New York: Cambridge University Press.

Pires, J. (1992) ‘Return and reintegration of qualified nationals from developing countries residing abroad: the IOM experience, International Migration, 30(3/4): 353-75.

33

Portes, A. (Ed) (1995) The Economic Sociology of Immigration. Essays on Networks, Ethnicity, and Entrepreneurship, New York: Russell Sage Foundation.

Portes, A. and Walton, J. (1981) Labour, Class, and the International System, New York: Academic Press.

Ranis, G. and Fei, J.C.H. (1961) ‘A theory of economic development’, American Economic Review, 51, 533-65.

Ravenstein, E.G. (1885) ‘The laws of migration’, Journal of the Royal Statistical Society, 48: 167-235.

Ricca, S. (1989) International Migration in Africa: Legal and Administrative Aspects, Geneva: International Labour Office.

Riccio, B. (2000) ‘Senegalese transnationalism: and ambivalent experience’, CDE Working Paper, www.sussex.ac.uk/units/CDE/level2/research/research.html.

Richmond, A.H. (1984) ‘Explaining return migration’, in D. Kubat (Ed) The politics of Return: International Return Migration in Europe, New York: Center for Migration Studies, 269-275.

Robin, N. (1996) Atlas des migrations ouest-africaines vers l’Europe, 1985-1993, Paris: Eurostat/ORSTOM.

Rogers, R. (1984) ‘Return Migration in comparative perspective’, in D. Kubat (Ed) The politics of Return: International Return Migration in Europe, New York: Center for Migration Studies, 277-299.

Rose, A.M. (1970) ‘Distance of migration and socio-economic status of migrants’, in C.J. Jansen (Ed) Readings in the sociology of migration, London: Pergamon Press, 85-92.

Russell, S.S. (1992) ‘Migrant remittances and development’, International Migration, 30(3/4), 267-87.

Russell, S.S. (1986) ‘Remittances from international migration: A review in perspective’, World Development, 14(6), 677-96.

Russell, S.S., Jacobsen, K. and Stanley, W.D. (1990) International Migration and Development

in Sub-Saharan Africa, World Bank Discussion Papers 101, Washington D.C.: World Bank.

Safa, H.I. and Du Toit B. (Eds) (1975) Migration and Development, The Hague: Mouton.

Saloutos, T. (1956) They Remember America: The Story of Repatriated Greek Americans, Berkeley: University of California Press.

Salt, J. and Findlay, A. (1989) ‘International migration of highly skilled manpower: theoretical and development issues’, in R. Appleyard (Ed) The Impact of International Migration on Developing Countries, Paris: OECD, 159-80.

Skeldon, R. (1997) Migration and Development: A Global Perspective, Longman.

Smith, M.P. and Guarnizo, L.E. (Eds) (1998) Transnationalism From Below, New Brunswick: Transaction Publishers.

Stahl, C.W. (1991) ‘South-North migration in the Asia-Pacific region’, International Migration, 23(2): 163-94.

Stahl, C.W. (1989) ‘Overview: economic perspectives’, in R. Appleyard The Impact of International Migration on Developing Countries, Paris: OECD, 361-87.

Stahl C. (Ed)(1988) International Migration Today, Volume 2: Emerging Issues, Paris: UNESCO.

Standing, G. (1984) ‘Income transfers and remittances’, in Migration Surveys in Low Income Countries: Guidelines for Survey and Questionnaire Design, R.E. Bilsborrow, A.S. Oberai, and G. Standing (Eds), London: Croom Helm.

Stinner, W.F., de Albuquerque, K. and Bryce-Laporte, R. (Eds) (1982) Return Migration and Remittances: Developing a Caribbean Perspective, Washington D.C.: Smithonian Institution.

Strachan, A.J. (1980) ‘Government sponsored return migration to Guyana’, Area, 12(2): 165-69.

Sutton, C. and Makiesky, S. (1975) ‘Migration and West Indian racial and ethnic counciousness’, in H.I. Safa and B. Du Toit (Eds) Migration and Development, The Hague: Mouton, 113-44.

34

Tapinos, G. and Delaunay D. (2000) ‘Can one really talk about the globalisation of migration flows?’, in Globalization, Migration, and Development, Paris: OECD, 35-58.

Taylor, E. (1976) ‘The social adjustment of returned migrants to Jamaica’, in F. Henry (Ed) Ethnicity in the Americas, The Hague: Mouton, 213-30.

Thomas-Hope, E. (1999) ‘Return migration to Jamaica and its development potential’, International Migration, 37(1): 183-207.

Todaro, M.P. (1989) Economic Development in the Third World, London: New York.

Todaro, M.P. (1976) International Migration in Developing Countries: A Review of Theory, Geneva: ILO.

United Nations Secretariat (1986) ‘The meaning, modalities and consequences of return migration’, International Migration Review, 23(1), 77-93.

Williams, T.O. (1999) ‘Factors affecting manure application by farmers in semi-arid West Africa’, Nutrient Cycling in Ecosystems, 55(1): 15-22.

World Bank (2000) Entering the 21st Century, World Development Report 1999/2000, New York: Oxford University Press.

Zachariah, K.C. and Condé, J. (1981) Migrations in West Africa: Demographic Aspects, Oxford: Oxford University Press.

Zachariah, K.C. (1980) ‘Migration in the Ivory Coast, in Demographic Aspects of Migration in West Africa, Vol. 2, Staff Working Paper 414, Washington D.C.: World Bank.

Zachariah, K.C. and Nair, N.K. (1980) ‘Demographic aspects of recent international and internal migration in Ghana, in Demographic Aspects of Migration in West Africa, Vol. 1, Staff Working Paper 415, Washington D.C.: World Bank.